LITERATURE REVIEW Public-Private Partnerships in … in... · 5 Public-Private Partnerships in...

Public-Private Partnerships in Basic Education: An International Review

Norman LaRocque

LITERATURE REVIEW


www.cfbt.com 2

Welcome to CfBT Education Trust

CfBT Education Trust is a leading charity providing education services for public benefit in the UK and internationally. Established 40 years ago, CfBT Education Trust now has an annual turnover exceeding £100 million and employs more than 2,000 staff worldwide who support educational reform, teach, advise, research and train.

Since we were founded, we have worked in more than 40 countries around the world. Our work involves teacher and leadership training, curriculum design and school improvement services. The majority of staff provide services direct to learners in schools or through projects for excluded pupils, in young offender institutions and in advice and guidance for young people.

Surpluses generated by our operations are reinvested in educational research and development. Our new research programme – Evidence for Education – will improve educational practice on the ground and widen access to research in the UK and overseas.

CfBT Education Trust is pleased to have commissioned this important survey of Public-Private Partnerships in education worldwide. As an education company CfBT has extensive experience in this field. Our work related to Public-Private Partnerships includes:

• Ghana and Kenya – working on behalf of the International Finance Corporation on Private Schools Support Programs

• Abu Dhabi – supervising school improvement in a number of government schools on behalf of the Abu Dhabi Education Council

• India – working with HSBC on support for budget private schools

• Dubai – providing consultancy on a school inspection approach for both government and private schools on behalf of the Knowledge and Human Developmen]t Authority

• Qatar – developing curriculum standards for state funded ‘independent’ schools on behalf of the Supreme Education Council.

Visit www.cfbt.com for more information.

Norman LaRocque is a public policy consultant and an advisor to the Education Forum, based in Wellington, New Zealand. His work covers all levels of education and has focused on the regulation of education, education finance, student loans, education contracting and public-private partnerships. Norman has worked in more than 15 countries in Asia and the Pacific, Africa and the Middle East. He has undertaken consulting projects for a range of

organisations, including the World Bank, the International Finance Corporation, GTZ, the Asian Development Bank and the IBM Center for the Business of Government. He is a Research Affiliate with the State University of New York at Albany’s Program on Private Higher Education.

Norman has an MA and a BA (Honors) in Economics from the University of Western Ontario (Canada).

About the Author

The author would like to thank Neil McIntosh, Tony McAleavy and Michael Latham from CfBT Education Trust for comments on an earlier draft of the paper. He would also like to thank the

Research and Development team, and the CfBT designers and editors for their excellent project coordination and editorial support. Any errors remain the author’s.

Acknowledgements

The views and opinions expressed in this publication are those of the authors and do not necessarily represent the views of CfBT Education Trust.

© CfBT copyright May 2008 All rights reserved

www.cfbt.com 3


Contents

1 Introduction 6

2 Public-Private Partnerships: Definition 7

3 International Examples of Innovative PPPs at the Basic Education Level 9

3.1 Private Sector Philanthropic Initiatives 9

3.2 Private Management of Public Schools 13

3.2.1 Contract Schools, United States 13

3.2.2 Charter Schools, United States 14

3.2.3 Colegios en Concesión, Colombia 15

3.2.4 Fe y Alegría, Latin America and Spain 17

3.2.5 Independent Schools, Qatar 17

3.2.6 School Management Initiatives, Pakistan 18

3.3 Government Purchase of Education Services from Private Schools 20

3.3.1 Universal Secondary Education Programme, Uganda 20

3.3.2 Government Sponsorship of Students in Private Schools, Côte d’Ivoire 20

3.3.3 Educational Service Contracting, The Philippines 20

3.3.4 Alternative Education, New Zealand 21

3.3.5 Venezuelan Association of Catholic Education Schools 22

3.3.6 Financial Assistance per Child Enrolled Basis Programme, Punjab (Pakistan) 22

3.4 Voucher and Voucher-like Programmes 22

3.4.1 Plan de Ampliación de Cobertura de la Educación Secundaria, Colombia 24

3.4.2 Targeted Individual Entitlement, New Zealand 24

3.4.3 Independent School Subsidies, New Zealand 25

3.4.4 School Funding System, The Netherlands 25

3.4.5 Milwaukee Parental Choice Program, Milwaukee, USA 25

3.4.6 Urban Girls’ Fellowship Program, Balochistan (Pakistan) 25

3.5 Adopt-a-School Programmes 26

3.5.1 Sindh Education Foundation 26

3.5.2 Adopt-a-School Programme, The Philippines 27

3.6 Capacity Building Initiatives 27

3.6.1 Cluster Based Training of Teachers, Punjab (Pakistan) 27

3.6.2 Quality Advancement and Institutional Development in Private Sector Schools 27

3.6.3 Quality Assurance Resource Centre 27

3.6.4 Teaching in Clusters by Subject Specialists 28

2www.cfbt.com 4


3.7 School Infrastructure Partnerships 28

3.7.1 Private Finance Initiative, United Kingdom 28

3.7.2 School Private Finance Projects, Australia 29

3.7.3 PPP for New Schools, Egypt 29

3.7.4 Public-Private Partnerships for Educational Infrastructure, Nova Scotia (Canada) 30

3.7.5 School Infrastructure PPPs, Germany and The Netherlands 30

3.7.6 P3 New Schools Project, Alberta (Canada) 30

3.7.7 Leasing of Public School Buildings to Private Operators, Pakistan 31

4 PPPs in Education: Lessons Learned and Future Directions 33

4.1.1 Evidence on the Impact of PPPs in Basic Education 33

4.1.2 The Role of Government in Education: Financing vs Provision 37

4.1.3 Issues and Lessons for PPP Design 39

4.1.4 Implications for Research and Parameters for Intervention Studies 47

5 Conclusion 50

6 Bibliography 51

Annexes 1 and 2 54

Table 1: Private Sector Share of Enrolments by Level of Education, Selected Countries, 2005 7

Table 2: Classification of PPPs at the Basic Education Level 10

Table 3: Total Giving by Area of Assistance and Nature of Assistance, Philippines, 2001 11

Table 4: Number of students in public schools managed by private providers, Colombia, 2004 17

Table 5: Voucher and Voucher-like Programmes Around the World 23

Table 6: Repetition Rates and Dropout Rates in FyA Schools and Other Public Schools 34

Table 7: Price Certainty and Project Delivery, Traditional Procurement Method vs PFI, UK 37

Table 8: Financing vs Provision in Government Services 37

List of Tables

www.cfbt.com 5


Figure 1: Number of, and Enrolments at, EMO Managed Schools, USA, 1998/99–2005/06 14

Figure 2: Number of Operating Charter Schools, United States, 1995/96–2007/08 15

Figure 3: Enrolments in Concession Schools, Bogotá, 2000–2004 16

Figure 4: Total number of students in Fe y Alegría schools, 1980–2005 18

Figure 5: Number of Independent Schools, Qatar, 2004/05–2007/08 19

Figure 6: Number of ESC grantees and participating schools, 1986/87–2005/06 21

Figure 7: Number of Participating Private Schools and Voucher Recipients, MCPC, 1990/91–2005/06 26

List of Figures

Box 1: Public-Private Partnerships Defined 8

Box 2: The Punjab Education Foundation 39

Box 3: Principles to Guide the Development of a Private School Subsidy Scheme 40

Box 4: Selected Characteristics of a Desirable PPP Framework 46

List of Boxes

Annex 1: Summary of Selected International Examples of PPPs at the Basic Education Level 54

Annex 2: Summary of Selected Evidence on Education PPPs 58

List of Annexes

2www.cfbt.com 6


1. Governments around the world, particularly those in developing countries, face significant educational challenges. According to the United Nations, about 115 million are not in school; the bulk of these children live in Sub-Saharan Africa and South Asia (United Nations 2005). While progress has been made toward meeting the education Millennium Development Goals (MDGs), much remains to be achieved. This is particularly true in the least developed countries.

2. In addition to a lack of access to schooling, the poor quality of education delivered means that most children who complete school in the developing world find they are not sufficiently prepared for the world of work. Today, education in the developing world faces the twin challenges of getting and keeping more children enrolled in school, while simultaneously ensuring that learning

outcomes improve. A number of governments have responded to these education challenges by making greater use of the private sector and public-private partnerships (PPPs) in education as a means of improving both the delivery and financing of Basic Education in developing and developed countries.

3. This report examines the international experience with PPPs at the Basic Education level. Several forms of PPP are highlighted, including private philanthropic initiatives, private sector management initiatives, private school funding programmes (e.g. subsidies and vouchers), adopt-a-school programmes, capacity building initiatives and school infrastructure partnerships. The report also draws a number of lessons for the design and implementation of PPPs, based on the review of international experience with PPPs.

1 Introduction

Today, education in the developing world faces the twin challenges of getting and keeping more children enrolled in school, while simultaneously ensuring that learning outcomes improve.

‘‘

‘‘

www.cfbt.com 7


4. Public delivery represents the norm at the Basic Education level in most developed and developing countries. Nonetheless, the private sector plays an important role in the delivery of Basic Education in many countries. The bulk of private participation at the Basic Education level in these countries involves the ‘traditional’ model of privately operated/privately financed schools (see Table 1). While private schools are often seen as catering solely to the wealthy,

the reality is that many countries – including Pakistan, India, Indonesia and several African countries – have seen the emergence of private schools charging modest fees that cater to students from low-income families.1 The private sector has also played an important role as a supplier of inputs, and to a lesser extent, as a provider of ancillary services such as school transport and food services through outsourcing arrangements.

TABLE 1: Private Sector Share of Enrolments by Level of Education, Selected Countries, 2005

Country Primary Level (%)

Secondary Level (%)

Lebanon 65.8 53.5

Australia 29.1 27.0

Indonesia 16.6 44.2

Philippines 7.6 19.9

Korea 1.3 34.2

Thailand 15.8 13.4

Chile 51.1 52.2

Colombia 18.7 23.8

Ecuador 28.8 33.4

Peru 16.4 22.2

Belgium 35.1 68.2

Denmark 12.1 13.0

France 15.1 25.0

Spain 33.4 28.2

UK 5.3 29.6

Pakistan 35.8 25.0

1 See for example Andrabi, Tahir et al (2006) A Dime a Day: The Possibilities and Limits of Private Schooling in Pakistan, World Bank Policy Research Paper No. 4066, World Bank, Washington DC and Srivastava, Prachi (2007) Neither Voice nor Loyalty: School Choice and the Low-Fee Private Sector in India, Research Publications Series, Occasional Paper No. 134, National Center for the Study of Privatisation in Education, New York.

2 Public-Private Partnerships: Definition

Source: UNESCO

While private schools are often seen as catering solely to the wealthy, the reality is that many countries…have seen the emergence of private schools charging modest fees that cater to students from low-income families.

‘‘

‘‘

2www.cfbt.com 8


5. Recent years have seen an expansion and broadening of the private sector’s role in the financing and provision of education services in many countries. A key trend has been the emergence of more sophisticated forms of private involvement in education through PPPs. PPPs involve the public and private sectors working together to achieve important educational, social and economic objectives. They represent a move away from the traditional model of government procurement for the delivery of public services. Despite the expansion of PPPs, and the increased attention they have received in recent years, there is little agreement about what constitutes a PPP or how they are defined (see Box 1). PPPs can be defined narrowly to include only formal arrangements such as sophisticated infrastructural initiatives or they can be defined more broadly to include all manner of partnership between the public and private sector.

6. Despite their broad scope, it is generally accepted that PPPs share a number of characteristics, including that they are formal in nature, involve the development of a long-term relationship between the partners, are outcome focused, include an element of risk-sharing among the partners and can

involve both the voluntary and commercial sectors as private sector partners. In all PPPs, the public sector’s role is essentially to define the scope of business; specify priorities, targets, and outputs; and set the performance regime by which the management of the PPP is given incentives to deliver. The essential role and responsibility of the private sector in all PPPs is to deliver the business objectives of the PPP on terms offering value for money to the public sector.

7. PPPs need to be contrasted with privatisation. As Wang (1999) notes, privatisation implies permanent transfer of control, whether as a consequence of a transfer of ownership right from a public agency to one or more private parties, or for example, of a capital increase to which the public sector shareholder has waived its right to subscribe. In contrast, PPPs aim to promote improvements in the financing and provision of services from both the public and private sectors but not to increase the role of one over the other. Rather, PPPs are geared toward improvement of existing services provided by both sectors with an emphasis directed on system efficiency, effectiveness, quality, equity and accountability.2

2 Wang, Yidan (1999) Public-Private Partnerships in Health and Education: Conceptual Issues and Options, paper prepared for Manila Social Policy Forum: The New Social Agenda for East, Southeast and Central Asia, Joint ADB-World Bank Conference, 9–12 November, pp. 6–7.

Box 1: Public-private Partnerships Defined

‘a risk-sharing relationship based upon an agreed aspiration between the public and private (including voluntary) sectors to bring about a desired public policy outcome. More often than not this takes the form of a long-term and flexible relationship, usually underpinned by contract, for the delivery of a publicly funded service.’

Commission on UK PPPs

‘a cooperative venture between the public and private sectors, built on the expertise of each partner, that best meets clearly defined public needs through the appopriate allocation of resources, risks and rewards.’

Canadian Council for PPPs

‘Arrangements whereby the private sector provides infrastructure assets and services that traditionally have been provided by government, such as hospitals, schools, prisons, roads, bridges, tunnels, railways, and water and sanitation plants.’

OECD

www.cfbt.com 9


8. The private sector has been involved in the delivery of ‘public’ services such as water and transport for many years. However, the extension of PPPs into social policy areas such as health and education is more recent and is arguably one of the most significant trends in public finance in the past decade. As will be highlighted below, there is a wide range of PPPs in use in the Basic Education sector – each with different characteristics, design features and country contexts. This section examines a number of examples of PPPs at the Basic Education level (see Table 2). For the purposes of this report, education PPPs can be classified into seven categories:

• private sector philanthropic initiatives;

• school management initiatives, under which education authorities contract directly with private providers to operate public schools or manage certain aspects of public school operations. Although these schools are privately managed, they remain publicly owned and funded;

• government purchase initiatives under which governments contract with private schools to deliver education at public expense;

• voucher and voucher-like initiatives under which governments fund students to attend private schools;

• adopt-a-school programmes under which private sector partners provide cash and in-kind resources to complement government funding of public schools;

• school capacity-building initiatives under which private sector partners provide teacher training and curriculum enhancement programmes; and

• school infrastructure initiatives under which private sector partners design, finance, construct and operate public school infrastructure under long-term contracts with the government.

9. Clearly, not all education PPPs fit neatly into this typology. First, different programmes may fall into different categories even where

they have broadly similar design features. For example, government purchase initiatives and voucher programmes are similar in many respects, although they do differ in that the former involve a more formal contracting arrangement between the government and private sector providers. Second, some programmes may contain elements that are common to different types of initiatives and the appropriate categorisation may depend on the way a programme is implemented and the nature of the regulation governing providers. These are discussed below.

3.1 Private Sector Philanthropic Initiatives

10. Perhaps the most common form of PPP in the Basic Education sector is private philanthropy. The United States has a long tradition of private philanthropy. In 2006, more than 70,000 private and community foundations in the United States disbursed grants totalling some $41 billion. Grants to the education sector represented 22.5 percent of all grants over $10,000 awarded by the 1,263 largest corporate and community foundations – second only to health (23 percent) and ahead of human services (13.8 percent) and arts and culture (12.2 percent). There are many examples of foundations providing funding to Basic Education. Since 2000, the Bill and Melinda Gates Foundation has invested more than $1.5 billion in the creation of high-quality, high-performing schools and systems. The Broad Education Foundation is a Los Angeles-based venture philanthropic organisation whose mission is ‘to improve urban K–12 public education through better governance, management, labor relations and competition’. Its flagship initiatives include financing the Broad Prize for Urban Education, the Broad Center for the Management of School Systems and the Broad Institute for School Systems. The Broad Foundation has also provided funding for the New Schools Venture Fund, a not-for-profit venture capital firm that raises capital and invests in education entrepreneurs

3 International Examples of Innovative PPPs at the Basic Education Level

For example, government purchase initiatives and voucher programmes are similar in many respects, although they do differ in that the former involve a more formal contracting arrangement between the government and private sector providers.

‘‘

‘‘

2www.cfbt.com 10


TABLE 2: Classification of PPPs at the Basic Education Level

PPP Initiative Examples

Private Sector Philanthropic Initiatives

• Philanthropic Foundations (USA, Philippines)• Academies Programme (UK)• Philanthropic Venture Funds (USA)• World Education Forum’s Global Education Initiative (Jordan, Egypt, India, Palestinian National Authority)

School Management Initiatives • Contract Schools (USA)• Charter Schools (USA and Alberta, Canada)• Concession Schools (Bogotá, Colombia)• Independent Schools (Qatar)• Private Management of Railways Schools (Pakistan)• Quality Education for All (Pakistan)• Management of Government Schools (Lahore, Pakistan)

Purchase of Educational Services from Private Schools

• Alternative Education (New Zealand)• Government Sponsorship of Students in Private Schools (Côte ’Ivoire)• Educational Service Contracting (The Philippines)• Fe y Alegría (South America/Spain)• Financial Assistance Per Child Enrolled Basis (Punjab, Pakistan)• Universal Post Primary Education and Training (Uganda)• Venezuelan Association of Catholic Education (Venezuela)

Adopt-a-School Programmes • Sindh Education Foundation• Adopt-a-School Programme, Philippines

Vouchers and Voucher-like Programmes

• Plan de Ampliación de Cobertura de la Educación Secundaria (Colombia)

• School Funding System (The Netherlands)• Targeted Individual Entitlement and Independent School Subsidies (New Zealand)

• Milwaukee Parental Choice Program (Milwaukee, USA)• Urban Girls’ Fellowship Program (Balochistan, Pakistan)

Capacity Building Initiatives • Cluster Based Training of Teachers Through PPP (Punjab, Pakistan) • Quality Assurance Resource Centre (Sindh, Pakistan)• Quality Advancement and Institutional Development (Sindh, Pakistan)• Teaching in Clusters by Subject Specialists (Punjab, Pakistan)

School Infrastructure Initiatives • P3 New Schools Project (Alberta, Canada)• Private Finance Initiative (United Kingdom)• New Schools’ Private Finance Project (New South Wales, Australia)• New Schools’ Public-Private Partnership Project (South Australia, Australia)

• PPP for New Schools (Egypt)• Public-Private Partnerships for Educational Infrastructure (Nova Scotia, Canada)

• Offenbach Schools Project and Cologne Schools Project (Germany)• Montaigne Lyceum (The Hague, The Netherlands)• Leasing of Public School Buildings to Private Operators (Punjab, Pakistan)

www.cfbt.com 11


whose initiatives serve minority and low-income students in under-served urban areas.

11. Philanthropic initiatives are by no means limited to the United States or to developed countries. India’s Bharti Foundation committed $50 million to the creation of strictly non-profit, private schools in the nation’s poor rural areas. Corporate foundations in the Philippines are well organised and donate considerable amounts to schools, both through the country’s Adopt-a-School programme and through various other initiatives. This work is coordinated by an umbrella group – the League of Corporate Foundations (www.lcf.org.ph) – which has developed a roadmap of corporate giving to the education sector. In 2001, corporate foundations in the Philippines donated $1,103,000 (cash and in-kind) to education causes (Table 3). This assistance included equipment, teacher training, buildings and instructional materials. Assistance to education represented 23 percent of total corporate giving in 2001.

12. In Pakistan, the Pakistan Centre for Philanthropy (PCP) plays a similar role to the LCF. It has a number of roles, including

providing support services to philanthropists and certifying not-for-profit organisations. A key objective is to increase the amount and effectiveness of corporate philanthropy to lift both access to, and the quality of, education for disadvantaged children. The PCP plays an important role in facilitating PPPs in the education sector. To this end, it has developed a framework for education PPPs. Corporate philanthropy in Pakistan is significant. A study carried out in 2000 showed that fully 93 percent of Pakistani companies engaged in some form of corporate philanthropy, with 35 percent involved in the education sector. This placed education second only to health. There are also well-established private initiatives that serve education including one of the most prominent, Aga Khan Education Services (AKES), which currently operates more than 300 schools and advanced educational programmes that provide quality pre-school, primary, secondary and higher secondary education services to students in Pakistan, India, Bangladesh, Kenya, the Kyrgyz Republic, Uganda, Tanzania and Tajikistan.

13. There is also a range of innovative ventures supporting Basic Education, both

TABLE 3: Total Giving by Area of Assistance and Nature of Assistance, Philippines, 2001

Area of Assistance Total Giving ($)

Percent of Total (Percent)

Cash Donations ($)

Cash Donations (Percent)

Education 1,103,571 23.0 819,643 74.3

Social Services 639,286 13.4 164,286 25.7

Organisational Support 539,286 11.2 500,000 92.7

Health 485,714 10.1 305,357 62.9

Culture and Arts 403,571 8.4 398,214 98.7

Support for Government Program 396,429 8.3 298,214 75.2

Livelihood/Communication Credit 278,571 5.8 232,143 83.3

Environment 242,857 5.1 180,357 74.3

Civic and Community Affairs 121,429 2.5 117,857 97.1

Disaster Relief and Rehabilitation 96,429 2.0 80,357 83.3

Other 478,571 10.2 380,357 79.5

Total 4,785,714 100.0 3,476,786 72.6

Source: Lozada-Crystal, D. (2001) RVR-AIM Report on Corporate Giving Philippines 2001, Manila, pp. 17–18.

2www.cfbt.com 12


philanthropic and for-profit. These include the World Economic Forum’s (WEF) Global Education Initiative (GEI), which aims to create new sustainable models for education reform in the developing world through PPPs. The first of these, the Jordan Education Initiative, was launched in 2003. Its objectives include improving the development and delivery of education in Jordan through PPPs, encouraging the development of an efficient public-private model for the acceleration of educational reforms in developing countries through the use of ICT and building the capacity of the Jordanian IT industry for the development of innovative learning solutions in partnership with world class firms. Similar initiatives have been launched in Egypt, in the Indian State of Rajasthan and with the Palestinian National Authority. In 2007, the GEI and UNESCO launched the Partnerships for Education programme, with the objective of creating a global coalition to advance multi-stakeholder partnerships in education to help progress the attainment of Education for All.

14. In February 2007, Orient Global, a Singapore-based private investment institution, announced the launch of a $100 million education fund which is investing in private education opportunities in developing countries, including the research and development for a low cost chain of schools. They are already supporting Sir Edmund Hillary’s Himalayan Trust education programme in Nepal and have invested $48 million to acquire a 9 percent share of NIIT Limited, India’s premier education and training company and a leading provider of global education services. NIIT trains 500,000 learners annually in over 3,300 education centres in 30 countries. Opportunity International, a leading innovator in the microfinance industry, recently announced the expansion of its microfinance school loans programme to bring greater educational opportunity to poor children, especially girls. The Microschools of Opportunity initiative provides loans to educational entrepreneurs who open schools in poor neighbourhoods where children cannot access public schools. Funds can be used to build latrines, refurbish classrooms or buy land. The Microschools initiative is being piloted in 50 neighbourhoods and towns in Ghana.

15. Corporate philanthropy is a key feature of the UK’s Academies programme. Academies are publicly funded independent schools that provide free education to students of all abilities. Academies are established by sponsors from business, faith or voluntary groups through partnerships with central Government and local education partners. The Government meets the capital and running costs for the Academy. Working with other partners, sponsors help to ensure that the school is properly designed and equipped to provide a high standard of education. They also make a significant contribution to the capital costs of the new or refurbished school building, with the Government providing the balance. Running costs are met by the Government. The first Academy opened in 2002. There are currently 47 Academies.

16. Philanthropic foundations partner with charter schools operated by not-for-profit charter management organisations (CMOs) such as Aspire Public Schools, Green Dot Public Schools and Leadership Public Schools. These foundations seek out high-quality charter operators and then work to replicate their successes elsewhere. The Charter School Growth Fund (CSGF) is a philanthropic venture fund founded to significantly increase the capacity of proven educational entrepreneurs to develop and grow networks of high quality charter schools. Through its activities, CSGF is enabling the national charter sector to grow more rapidly and provide a better quality of education to more children than is possible through the growth of single schools. Nationally, the CSGF will increase the number and scale of high quality educational options through the development and expansion of charter school management and support organisations. With a pool of $100 million, the CSGF expects to create 100,000 new, permanent places for underserved families in high quality charter schools by the year 2015.

17. The New Schools Venture Fund (NVF) is a venture philanthropy firm founded in 1998 that supports education entrepreneurs and connects their work to wider systems change. Through its first two funds, NVF invested more than $70 million in 30 entrepreneurial educational ventures. Its third fund will focus on fueling the growth and quality of charter

Academies are publicly funded independent schools that provide free education to students of all abilities.

‘‘ ‘‘

www.cfbt.com 13


schools and on supporting the people, tools and practices needed for public school systems to become performance driven organisations. NVF raises capital from individual and institutional investors and uses those funds to support promising education entrepreneurs — particularly within targeted urban areas. In 2003, NVF received a grant from the Bill & Melinda Gates Foundation to create at least five CMOs that together will establish 20 new schools serving some 8,000 students within the first two years of operation. By their tenth year, each organisation is expected to launch 20 schools, for a total of 100 new schools serving 40,000 students. A subsequent grant in 2006 is supporting up to 20 additional entrepreneurial charter school developers. The grant will help establish nearly 200 high-quality schools, ultimately serving 100,000 students in a number of low-income urban communities.

3.2 Private Management of Public Schools

18. The first form of PPP in the education sector is the private management of public schools. This involves education authorities contracting directly with private providers to operate public schools or certain aspects of public school operations. While these schools are privately managed, they remain publicly owned and publicly funded.

19. Contract schools can be run by a variety of bodies – including private firms, neighbouring schools with a good reputation for serving students and their community, non government organisations (NGOs), universities, etc. Contract schools are individual legal entities capable of negotiating contracts, spending public funds and hiring and firing staff. They also have the legal authority to defend their interests in court against the funding authority. Contracts contain basic requirements that apply to private schools but also outline expected student outcomes, methods for assessing those outcomes, the goals of the school, and its programme of instruction. The contract also covers the agreed or mandated curriculum. As part of the contract, the management company or organisation is generally required to meet specific benchmarks in areas such as school

attendance, student performance and community involvement.

20. There are a number of examples of contract schools in both developed and developing countries, including Colombia, Qatar and the United States. Examples of these programmes are discussed below.

3.2.1 Contract Schools, United States

21. The private management of public schools in the United States has existed since the early 1990s and can take either of two forms. The first involves local school boards contracting directly with an Education Management Organisation (EMO) to manage a public school. The second involves indirect contracting where EMOs manage charter schools either as the holder of, or under contract to, the organisation that holds the school charter. Contract schools remain publicly owned and funded. Students do not pay fees to attend these schools. Typically, private sector operators are brought in to manage low performing schools in a given school district.

22. Private sector school operators may operate under either ‘management contracts’ or ‘operational contracts’. Under the former, the management of the school is turned over to the private sector; but teaching and other staff remain employed by the local school board and are subject to the teacher union contract. Under the latter, teaching and other staff are employed by the private operator, and terms and conditions of employment may differ from the teacher union contract. Under both of the above models, the private sector operator is paid a fixed amount per student (usually equal to the average cost of educating a student in the public sector) or is paid a fixed management fee and must meet specific performance benchmarks.

23. In 2005/06, there were 521 public schools managed by 51 companies in 29 states and the District of Columbia (see Figure 1). Eighty-four percent of these were fully privately managed charter schools (down from 86 percent in 2004/05). In 2005/06, there were over 237,000 students in privately managed public schools. The

Contract schools can be run by a variety of bodies – including private firms, neighbouring schools with a good reputation for serving students and their community, NGOs, universities, etc.

‘‘

‘‘

2www.cfbt.com 14


largest EMOs were Edison Schools and National Heritage Academies.3

24. The most significant example of private management of public schools is in Philadelphia, where a state takeover of the city’s schools resulted in 70 low performing ones being contracted out to for-profit and not-for-profit private providers in 2002. The largest provider, for-profit EMO Edison Schools was awarded management contracts for 20 schools. This was increased to 22 in 2005. Other examples include Denver Public Schools, which has used a small number of contract schools since 1993 and Chicago Public Schools, which is contracting out the management of a number of its lowest performing public schools as part of its Renaissance 2010 initiative.

3.2.2 Charter Schools, United States

25. Charter schools are secular public schools of choice that operate with freedom

from many of the regulations that apply to traditional public schools, such as geographic enrolment restrictions and teacher union contracts. The charter that establishes a school is a performance contract that details the school’s mission, programme, goals, students served, methods of assessment and ways in which success will be measured. Charter schools may be managed by the community or by a for-profit or not-for-profit school manager.

26. School charters may be granted by a district school board, university or other authorising agency. The term of a charter can vary, but most are granted for three to five years. Charter schools are accountable to their sponsor or authorising agency to produce positive academic results and adhere to the charter contract. A school’s charter can be revoked if guidelines on curriculum and management are not followed or standards are not met. At the end of the term of the charter,

600

500

400

300

200

135

230

285

368

406

463

535521

100

01998/99 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06

50,000

0

100,000

150,000

200,000

250,000

300,000

Num

ber

of s

cho

ols

Enr

olm

ents

Schools under EMO Management Enrolments

Source: Molnar et al (2006), p 2.

FIGURE 1: Number of, and Enrolments at, EMo Managed Schools, USA, 1998/99–2005/06

3 Molnar, Alex et al (2006) Profiles of For-Profit Education Management Organizations: 2005–2006, Eighth Annual Report, Education Policy Studies Laboratory, Arizona State University, Tempe, Arizona, p. 2.

www.cfbt.com 15


the entity granting the charter may renew the school’s contract. The advantage for charter schools’ increased autonomy is strengthened accountability.

27. The first charter school law was passed in the State of Minnesota in 1991, with the first charter school opening the following year. In the 2007/08 school year, there were more than 4,100 charter schools serving over 1.2 million students in the United States. The number of charter schools has increased considerably since the mid-1990s (see Figure 2). More than 40 states have passed charter school laws.

28. A recent trend has been the development of CMOs – nonprofit networks of schools that serve a specific geographic area. The CMO model seeks to maximise quality and sustainability, while providing scope for scaling up the charter school model within a targeted geographic area. By centralising or sharing certain functions and resources across schools, CMOs may offer a more viable model for wholesale reform of large public school systems than does a system of individual charter schools. Philanthropic foundations such as the Bill and Melinda Gates Foundation and NVF are involved in the development of CMOs.

3.2.3 Colegios en Concesión, Colombia

29. In Colombia, the City of Bogotá has introduced the Colegios en Concesión (Concession Schools) programme, under which the management of some public schools is turned over to private schools with proven track records of delivering high-quality education. The first Concession Schools began operating in 2000. In 2004, there were 25 schools, serving over 26,000 students. The programme was expected to grow to approximately 45,000 students in 51 schools (about 5 percent of public school coverage in Bogotá). However, a change in the mayoralty has reportedly stopped the programme’s expansion.

30. The Concession Schools programme is designed to overcome many of the traditional problems faced by public schools, including weak leadership, the inability of schools to select their own personnel, lack of labour flexibility, lack of equipment and supplies, bureaucratic red-tape and the politicisation/unionisation of the education sector.

31. Under the Concession Schools programme, private schools and/or educational organisations bid in a competitive process for management contracts of newly

The Concession Schools programme is designed to overcome many of the traditional problems faced by public schools, including weak leadership…

‘‘ ‘‘

1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08

4500

4000

3500

3000

2500

2000

1500

1000

500

500

253441

733

1,208

1,684

2,069

2,371

2,695

2,970

3,4003,625

3,9404,147

Num

ber

of C

hart

er S

cho

ols

Source: Vanourek (2005) and Center for Education Reform

FIGURE 2: Number of operating Charter Schools, United States, 1995/96 to 2007/08

2www.cfbt.com 16


built schools in poor neighbourhoods of Bogotá. Contractors may manage a single school or a group of schools. The schools must provide educational services to children who are poor and are paid Col$1,114,500 per full-time student per year – an amount that is considerably below the average cost of a student who attends a public school for only a half day (see Figure 3.)

32. Management contracts are for 15 years, which demonstrates both long-term commitment to educational improvement and continuity in supply. Contracts with providers establish clear standards that must be met, including hours of instruction, quality of nutritional provision and the establishment of single shift schooling. The provider has full autonomy over school management and is evaluated on results. Contracts with providers are performance based. Failure to meet educational outcome targets, such as standardised test scores and drop-out rates for two consecutive years, can result in the cancellation of the contract.

33. School inspections are carried out by a private firm to monitor the maintenance of school facilities and property. In addition, the Bogotá Ministry of Education carries out ongoing reviews of pedagogical standards and norms, and finances an independent evaluation to determine whether academic objectives have been met. While it remains early days for the programme, initial results show it has led to some management improvements, including a reduction in the share of the budget allocated to human resources, which has released money for nutritional support and the purchase of textbooks and teaching materials.4 Educators have also expressed satisfaction with the increased level of autonomy that the schools enjoy. There is a high demand for more such schools among the local community.

34. Other forms of school contracting exist throughout Colombia, including in Medellín and Cali. Contract schools are known as ‘schools under bidding’ in Medellín, as ‘management modality’ in Cali and ‘managed

4 Rodriguez, Alberto and Kate Hovde (2002) The Challenge of School Autonomy: Supporting Principals, LCSHD Paper Series No 77, Department of Human Development, The World Bank, Washington DC.

Source: Secretaría de Educación de Bogotá.

30,000

7,007

25,000

20,000

15,000

10,000

5,000

Enr

olm

ents

0

2000 2001 2002 2003 2004

FIGURE 3: Enrolments in Concession Schools, Bogotá, 2000–2004

19,374

21,430

23,117

26,308

School inspections are carried out by a private firm to monitor the maintenance of school facilities and property.

‘‘ ‘‘

www.cfbt.com 17


and directed schools’ in the district of Cartagena.5 In 2004, nearly 70,000 students were in privately managed public schools in Colombia (see Table 4).

3.2.4 Fe y Alegría, Latin America and Spain

35. FyA is a Jesuit controlled NGO that operates formal pre-school, primary, secondary and technical education programmes in the poorest communities in Latin America and Spain. The programme began in Venezuela in 1955 and has since spread to 14 other countries. FyA’s primary mission is to provide quality education to poor people, to ensure that students complete at least the basic cycle of schooling, and to establish schools that operate on behalf of community development. Under the FyA model:

• ministries of education pay the salaries of teachers and the principal;

• foundations, international agencies and voluntary fees from the local community pay for the land, construction and maintenance of schools;

• the community invites FyA to open a school and build it; and

• FyA trains and supervises teachers, manages the school and assists it in its operation as a community development centre.

36. A national office coordinates the network of FyA schools in each country, while overall coordination is provided by headquarters in Venezuela. Most FyA schools are located in rural areas, but some are found in or near urban slums. FyA schools can be either public or private, although a majority are public. Schools generally enjoy considerable autonomy – they can appoint school directors and teachers without state or teacher union interference. The central curriculum is supplemented with locally developed materials. FyA schools do not charge compulsory fees. The main indicator of school performance is student retention.6

37. In 2005, there were more than 1.2 million students in the FyA network – up from just 220,000 in 1980 (see Figure 4). Over 500,000 of these FyA programme participants were in formal education programmes.

3.2.5 Independent Schools, Qatar

38. Independent Schools are government-funded schools that have more operational autonomy and flexibility than regular public

5 World Bank (2005) Colombia: Contracting Education Services, Report No 31841-CO, The World Bank, Washington, DC, p 33.

6 McMeekin, Robert W (2003) Networks of Schools, Education Policy Analysis Archives, Arizona State University, Tempe, Arizona.

TABLE 4: Number of Students in Public Schools Managed by Private Providers, Colombia, 2004

Territorial Entity Number of Students Enrolled (Number)

Share of Total Public Enrolments (Percent)

Bogotá 25,589 3.3

Medellín 21,073 5.6

Cali 11,551 6.4

Cartagena 3,010 1.9

Valle del Cauca 4,000 1.1

Cundinamarca 3,000 0.7

Total 69,223 3.0

Source: World Bank (2005)

2www.cfbt.com 18


schools. They are run by government-selected operators and are overseen by the Supreme Education Council (SEC). Initially, school operators could be from a range of backgrounds, but the government has since changed the rules and now requires operators to also be the school’s principal. Independent Schools may be either newly established or converted existing Ministry of Education (MoE) schools.

39. Independent Schools are government-funded based on the number of students enrolled. Schools cannot charge tuition to Qataris and others eligible for public education, but they can charge tuition to students who are not eligible for a subsidy. Tuition fees cannot exceed the per-student subsidy provided by the government. Schools are free to set their own personnel policies and hire the staff they want. Staff in existing MoE schools do not retain their positions when a school converts to independent status. Class sizes cannot exceed 25 in Independent Schools.

40. Operators are granted a 3-year agreement to run an Independent School, and may renew for a further three years if

performance is satisfactory. The operating agreement describes the rights and responsibilities of each party and authorises disbursement of funding. Schools are encouraged to operate open admission policies, but may select students with the approval of the SEC. Independent Schools must produce an annual report and are evaluated by the Evaluation Institute. Operators are allowed to make a reasonable profit from the venture.

41. The Independent Schools programme was initiated in 2004/05. There are currently 64 Independent Schools operating in Qatar (see Figure 5 on page 19). In early 2006, the government announced that all MoE schools would become independent by 2010 or 2011.

3.2.6 School Management Initiatives, Pakistan

42. A number of school management initiatives are in place in the Punjab province of initiatives are in place in Pakistan. First, in the Punjab Province, Cooperation for Advancement, Rehabilitation and Education (CARE) – a local NGO – manages 170 government schools in Lahore City and two

1,400,000E

nrol

men

ts

1,200,000

1,000,000

800,000

600,000

400,000

200,000

–

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

FIGURE 4: Total Number of Students in Fe y Alegría Schools, 1980–2005

Source: www.feyalegria.org

www.cfbt.com 19


in Sarghoda, with enrolments of 100,000 students. Under this model, a CARE teacher is appointed as an internal co-ordinator (IC) at each school, acts as a supervisor and monitors the performance of both CARE and government teachers. The IC works in tandem with the government headteacher and is supervised weekly by an external coordinator (EC). CARE also hires an academic co-ordinator (AC) to supervise the functioning of the school. The EC and AC monitor the school to ensure regular attendance of teachers and students, monitor performance of government and CARE staff, ensure that student work is checked properly and that tests are administered and recorded regularly. CARE does not have direct administrative control over government staff, who remain public employees. The government pays the salaries of its 2,000 teachers, while CARE employs and pays a further 1,000 teachers. CARE does not receive any funding from the government beyond the payment of government teachers’ salaries.

43. A second initiative operated from 2003–2005. Under that model, Pakistan Railways contracted Beaconhouse Schools

to manage the 19 schools (with 13,850 students) for the children of employees. These schools were found in several locations around Pakistan, including Karachi, Risalpur, Faisalabad, Sukkur and Lahore. Pakistan Railways part-funded teachers’ salaries in the first 3 years of the contract. Schools charged a very low fee of only Rs25 per student per month, but they were allowed to enrol fee-paying private students to help offset the cost of operating the schools. Beaconhouse Schools did not have control over schools’ personnel policies (i.e. salaries, hiring and firing). The management contract was for 33 years, but it was ended after three years.

44. A third example, involves the Quality Education for All project of the National Rural Support Programme (NRSP), a semi-autonomous not-for-profit agency, which was contracted by the district government of Rahim Yar Khan in the Punjab to take over the management of 48 rural schools in a union council in July 2002. The purpose was to improve the quality of education in primary schools, reduce the number of drop-outs and increase enrolments. Since its inception, the programme has been

2004/05 2005/06 2006/07 2007/08

Num

ber

of S

cho

ols

70

60

50

40

30

20

10

0

12

33

46

64

FIGURE 5: Number of Independent Schools, Qatar, 2004/05–2007/08

Source: www.english.education.gov.qa

2www.cfbt.com 20


expanded to more than 2,400 schools. The programme was modelled on a similar one in the health sector under which the Punjab government established a system of upgraded Basic Health Units in rural areas. Under this arrangement, the entire operational budget for the schools was transferred to NRSP, together with responsibility for maintenance and authority over staff. The government remains responsible for capital works. The PPP is governed by a memorandum of understanding – effectively a management contract – which sets out performance targets and accountabilities. It also sets out arbitration clauses in case of dispute.

3.3 Government Purchase of Education Services from Private Schools

3.3.1 Universal Secondary Education Programme, Uganda

45. In February 2007, the Government of Uganda introduced a policy of Universal Secondary Education (USE). Under the policy, the government pays a subsidy for each student enrolled in eligible private secondary schools. Participation in the USE programme is limited to private secondary schools in sub-counties that are not served by government-aided or public schools. Only schools charging no more than Ush75,000 per student per term can participate in the programme. Participating schools receive a subsidy of Ush47,000 per student per term – well below what participating government-aided schools receive.

46. Participating schools are chosen by Ministry of Education and Sports. A Memorandum of Understanding is signed with individual private schools to ensure compliance with the policy’s implementation guidelines for private schools. In 2008, there are some 430 private secondary schools – serving approximately 56,000 students – participating in the USE programme. These figures are up from 363 schools and 42,000 students in 2007.

3.3.2 Government Sponsorship of Students in Private Schools, Côte d’Ivoire

47. The number of places available in public schools and training institutions in Côte d’Ivoire is insufficient to meet student demand. In addition, gross and net enrolment ratios in Côte d’Ivoire are low, even by Sub-Saharan Africa standards. To help bridge the gap in the supply of places, the government has introduced a programme whereby it sponsors ‘public’ students to attend private institutions. Under the programme, private schools receive a payment for each ‘public’ student they enrol. The government sponsors students in lower and upper secondary school and in professional and technical training. Students can be sponsored to attend both religious and secular schools.

48. The payment amount varies with the student’s educational level: $200 per year for lower secondary students and $233 per year for upper secondary students. Schools must be ‘chartered’ to take on sponsored students and the placement of students depends in part on the educational performance of the school. The number of students in the private school sponsorship programme grew from 116,000 in 1993 to 223,000 in 2001, an increase of 92 percent.7

3.3.3 Educational Service Contracting, Philippines

49. The Educational Service Contracting (ESC) scheme in the Philippines was introduced as a pilot in the early 1980s and made permanent in the late 1980s. Under ESC, the government contracts with private schools to enrol students in areas where there is a shortage of places in public high schools. The per-student payment to private schools can be up to PhP4,000 and cannot exceed the unit cost of delivery in public high schools.

50. Assistance under the Government Assistance to Students and Teachers in Private Education (GASTPE) is generally restricted to students at institutions that

7 Patrinos, Harry Anthony (2005) Education Contracting: Scope of Future Research, Program on Education Policy and Governance Report 05–23, Harvard University, Cambridge MA, p 7.

The payment amount varies with the student’s educational level: $200 per year for lower secondary students and $233 per year for upper secondary students.

‘‘ ‘‘

www.cfbt.com 21


charge low fees, and preference is generally given to students from low-income families. The GASTPE scheme is administered by the Fund for Assistance to Private Education (FAPE), a private not-for-profit organisation. The Department of Education (DepEd) recently introduced a certification programme for schools participating in ESC, which aims to address concerns about the quality of education at some schools.

51. In 2005/06, over 380,000 students in 1,833 participating private schools were subsidised under the ESC programme (see Figure 6). Grantee numbers and participating schools are up from just 4,300 and 158 respectively in 1986/87. The number of grantees and participating schools are up considerably since 2003/04.

3.3.4 Alternative Education, New Zealand

52. The Alternative Education (AE) programme in New Zealand funds the delivery of education in non-school settings for school-age children who have become alienated from the education system. The programme, which

was introduced in 1997, aims to give students a learning pathway to prepare them to return to mainstream secondary education or to move on to tertiary education or employment once they reach 16 years (or 15 years if they are granted an exemption from compulsory school attendance rules).

53. The management and delivery of AE can vary depending on local needs. A single school may contract for AE provision or groups of schools may form a consortium. School consortia vary in size. Students must be enrolled at a school in order to participate in AE. The AE programme may be delivered on or off the school site and schools may deliver the programme themselves or contract providers to offer the AE programme. Schools are responsible for the quality of AE programmes delivered by providers and for the students’ educational outcomes. Off-site programmes may be delivered by not-for-profit/community-based organisations or by for-profit educational providers. Schools are funded for AE on a per-student basis through contracts with the Ministry of Education (MoE). Schools sign a Memorandum of Understanding with the MoE

450,000

Num

ber

of G

rant

ees

Num

ber

of P

artic

ipat

ing

Sch

oo

ls

400,000

350,000

300,000

250,000

200,000

150,000

100,000

50,000

0

2,000

1,800

1,600

1,400

1,200

1,000

800

600

400

200

0

1986

/87

1987

/88

1988

/89

1989

/90

1990/

91

1991

/92

1992

/93

1993

/94

1994

/95

1995

/96

1996/

97

1997

/98

1998

/99

1999

/00

2000/

01

2001

/02

2002

/03

2003

/04

2004

/05

2005

/06

ESC Grantees Participating Schools

FIGURE 6: Number of ESC Grantees and Participating Schools, 1986/87 to 2005/06

Source: FAPE and DepEd data.

2www.cfbt.com 22


detailing each party’s responsibilities. In 2003, over 3,100 students were enrolled in AE at some time. There were 200 AE providers in 2004 – up from approximately 120 in 2001/02.8

3.3.5 Venezuelan Association of Catholic Education Schools

54. The Venezuelan Association of Catholic Education (AVEC) runs over 700 Catholic schools, most of which deliver education to poor children. In 1990, the Ministry of Education, Culture, and Sport (MECD) struck an agreement with AVEC to provide subsidies to private schools located in low-income urban and rural areas, indigenous communities, vocational schools, and those schools that were only able to cover up to 85 percent of operational costs. Fe y Alegria schools form a large body of AVEC schools. In 2005, government subsidies to AVEC schools amounted to almost $49 million, covering 483,000 students.

55. The agreement between MECD and AVEC holds AVEC schools to a higher degree of accountability for performance than public schools.9 Whereas there are no conditions imposed on public schools, AVEC schools must provide financial statements to MECD regarding the use of funds and present an annual management report. Supervision is also a large part of the AVEC model: supervisors visit the schools twice a year to assess the academic and operational situation. This information is then compiled for future analysis.

3.3.6 Financial Assistance per Child Enrolled Basis Programme, Punjab (Pakistan)

56. The Punjab Education Foundation (PEF) in the Punjab province of Pakistan operates several funding-based PPP programmes. The largest – the Financial Assistance per Child Enrolled Basis (FAS) programme – was established in late 2005. It has grown quickly, with programme coverage increasing from just 20,000 students in 54 schools in late 2005 to more than 500,000 students in 1,157 schools today. The programme pays participating

private schools Rs300 per month per enrolled student. Schools cannot charge fees on top of the per-student subsidy paid by the PEF. Schools must have at least 100 students and the maximum enrolment cannot exceed 750. Subsidies are paid directly to the school.

57. Assistance is focused on rural and poor areas. At present, FAS partner schools are located in a number of districts, including Lahore, Khushab, Bahawalpur, Chakwal, Bhakkar, Bahawalnagar and Sialkot. Schools must apply to participate in the programme and must enter into a partnership agreement with the PEF. PEF officials are given unrestricted access to partner schools to monitor enrolments, attendance, physical facilities and infrastructure and to conduct Quality Assurance Tests to gauge learning outcomes and quality standards. Students at FAS partner schools must continue to meet minimum performance benchmarks in order for the school to remain part of the FAS programme and hence eligible for funding. The PEF provides professional development support for partner FAS schools. The PEF has introduced other funding-based PPP programmes, including a small voucher scheme that operates in slum areas of Lahore.

3.4 Voucher and Voucher-like Programmes

58. A school voucher is a certificate or entitlement that parents can use to pay for the education of their children at a public or private school of their choice, rather than the public school that is closest to them or to which they have been assigned. Vouchers are paid directly from a public entity to parents or to schools directly on parents’ behalf.

59. There are many examples around the world – in both developed and developing countries – of voucher programmes that provide funding to students attending either public or private schools (see Table 5). These programmes may have different objectives – for example, improving quality

8 Education Review Office (2004) Alternative Education Report, June, ERO, Wellington.9 Mora, J.G. (2005) Public-Private Partnerships in Latin America: A Review Based on Four Case Studies, Paper presented at the Conference on Mobilizing the Private Sector for Public Education, 5–6 October, Harvard University, Cambridge, MA.

www.cfbt.com 23


TABLE 5: Voucher and Voucher-like Programmes Around the World

Country Mechanism

Bangladesh Stipends for girls to attend public or private schools

Belize Government partnerships with churches to share costs

Bolivia Private management (church-based organisation) of public schools

Brazil Matching grants, capitation grants, scholarships for poor students

Botswana Matching-grant schemes

Chad Community financing

Chile Voucher system for poor students, capitation grants for all students

China Matching-grant schemes, targeted bursary for poor and minority children

Colombia Targeted voucher system

Côte d’Ivoire Government sponsorship of students at private institutions

Dominican Republic Assistance to private schools serving low-income students

El Salvador School choice for poor

Gambia Targeted scholarships, capitation grants for all students

Guatemala Targeted stipends for girls in 13 communities

Ghana Matching-grant schemes

India Matching-grant schemes and numerous incentives

Indonesia Targeted scholarships for junior secondary school students

Jamaica Student loans

Kenya Voucher for informal sector workers for short-term skill upgrading courses

Lesotho Government partnership with churches to share costs

Mauritius Matching-grant schemes

Mexico Targeted bursary for poor and indigenous populations

Myanmar Community-sponsored schools

Morocco Scholarships for rural girls

Mozambique Scholarships for rural girls

Pakistan Community scholarships, subsidies to private schools serving rural girls

Senegal Scholarships for students to attend private and public schools in Dakar

Tanzania Matching-grant schemes, targeted bursaries for secondary school girls

Thailand Bicycles for poor students in rural areas

Zimbabwe Per capita grants

Source: Patrinos, H.A. (2000) ‘Market Forces in Education’. European Journal of Education 35(1), pp. 61–80.

2www.cfbt.com 24


and/or increasing educational access either generally or for specific groups. Voucher programmes may also have radically different design features and associated rules and regulations relating to eligibility, fee charges, school registration and student admissions. For example, some programmes may allow schools to charge fees on top of the value of the voucher, while others may not. Several examples of voucher schemes are examined below. Many others could be reviewed, including those in Denmark and Sweden.

3.4.1 Plan de Ampliación de Cobertura de la Educación Secundaria, Colombia

60. In the 1990s, Colombia experimented with a targeted voucher programme whose objective was to increase access to secondary schooling for children from poor families. The Plan de Ampliación de Cobertura de la Educación Secundaria (PACES) provided 125,000 vouchers during its six years in operation from 1992 to 1997. The programme offered vouchers to students entering sixth grade, the start of Colombian secondary school. Key elements of the programme included:

• vouchers were available to children from low-income families who had been attending a public primary school and who had been accepted at a private school;

• vouchers were renewable subject to satisfactory academic performance;

• the value of the voucher was $US190 – about half the cost of attending a private secondary school;

• the voucher was deposited by the student and the school received funds directly from the bank, rather than an educational intermediary;

• schools were allowed to charge top-up fees; and

• there was minimal regulation of private schools.10

61. The programme was successful in many respects. As noted by Patrinos (2005), the programme led to considerable enrolment increases, especially for the disadvantaged, at a low cost to government. The quality of education provided under the programme was at least comparable to that provided in public schools, yet the per beneficiary cost of the programme was about 77 percent of the unit cost of public secondary education.11

3.4.2 Targeted Individual Entitlement, New Zealand

62. The Targeted Individual Entitlement (TIE) programme was introduced in 1996 as a three-year pilot scheme. In 1998, it was given indefinite funding. The TIE programme was designed to assist children from low-income families to attend a private school, to give choice to families whose education options were limited and to lift educational achievement among low-income families. Under the scheme, the Government funded a small number (160) of children per year to be educated in private schools. Private schools received 110 percent of the average cost of education at a state school for each TIE student accepted. Families also received an allowance to cover non-tuition costs.

63. To be eligible for the TIE programme, students had to come from families with an annual income below a threshold level. Students were eligible to receive funding for up to six years. Participating schools had to be registered and had to offer the National Curriculum. The application process was managed by the organisation representing independent schools in New Zealand. An evaluation conducted in the late 1990s concluded that the programme was successful in increasing access to private schooling for children from low-income families and that parents, students and schools were highly satisfied with the programme. Despite this, the TIE scheme was abolished in 2000 and no new students were allowed into the programme from that year.12

10 Bettinger, Eric (2005) Lessons from Private-School Vouchers in Colombia, Federal Reserve Bank of Cleveland, Cleveland, www.clevelandfed.org/research/EdConf2005/Papers/BettingerPaperFINALweb.pdf.

11 Patrinos, Harry Anthony (2005) Education Contracting: Scope of Future Research, PEPG Report 05–23, Program on Education Policy and Governance, Harvard University, Cambridge Massachusetts.

12 Gaffney, Michael and Anne B. Smith (2001) ‘New Zealand’s Targeted Individual Entitlement Scheme’, in Can the Market Save Our Schools? Fraser Institute, Vancouver, pp. 151–166.

To be eligible for the TIE programme, students had to come from families with an annual income below a threshold level.

‘‘ ‘‘

www.cfbt.com 25


3.4.3 Independent School Subsidies, New Zealand

64. In New Zealand, independent schools receive government subsidies that are estimated at 25 to 35 percent of the average per pupil cost of educating a child in a government school. Subsidies are enrolment-based and vary with grade level. Independent schools are privately owned and largely privately funded. In order to be eligible to receive subsidies, schools must be registered. Beyond that requirement, there is minimal regulation of independent schools. For example, independent schools can be for-profit or not-for-profit, are free to charge top-up fees, are not part of the national teachers’ contract and are not required to teach the New Zealand curriculum (although most do). The independent school sector makes up about 4 percent of school-level enrolments and approximately 25,000 students are subsidised each year.13

3.4.4 School Funding System, The Netherlands

65. The Dutch education system has been decentralised and demand-driven since 1917. Almost 70 percent of schools in the Netherlands are administered and governed by private school boards. Public and private schools are funded by the government on an equal footing, and most parents have a choice of several schools near their homes. Parental choice has spurred some schools to develop a unique profile and to improve the education they offer. Schools are free to determine what is taught and how, the Ministry of Education does impose a number of statutory quality standards. The Education Inspectorate supervises schools.

66. In recent years, there has been a trend towards greater autonomy and decentralisation. Many central government powers have been transferred to the level of the individual school. The central government’s role is increasingly confined to broad policy-making and to creating the right conditions for

the provision of quality education. Institutions are being given greater freedom in the way they allocate their resources and manage their own affairs, although they still answer to government for their performance and policies. Schools receive extra funds to combat educational disadvantage. Additional funding is provided for schools in districts and regions with high numbers of underprivileged families.14

3.4.5 Milwaukee Parental Choice Program, Milwaukee, USA

67. The Milwaukee Parental Choice Program (MCPC) was established in 1990 in the City of Milwaukee in the US state of Wisconsin. It provides vouchers to poor families to allow them to send their children to private schools at state expense. Vouchers are available for students from kindergarten through grade 12 and are valued at up to $6,351 per year. To be eligible for a voucher, students must reside in Milwaukee and live in a household with an income equal to or less than 175 percent of the federal poverty level (220 percent for students or siblings of students who participated in the MPCP in the previous school year).

68. The state Department of Public Instruction maintains a list of participating private schools. Participating private schools may be secular or religious and must be accredited. They must also administer a nationally normed test in reading, math and science at various grade levels and cannot charge additional tuition fees on top of the voucher amount, although they can charge other fees for social or extra-curricular activities. The number of vouchers available in Milwaukee is capped. For 2006/07, the cap is 22,500 vouchers – up from 15,000 the previous year. There were 121 participating private schools and approximately 15,000 voucher recipients in 2005/06 (see Figure 7).

3.4.6 Urban Girls’ Fellowship Program, Balochistan (Pakistan)

69. The Urban Girls’ Fellowship Program (UGF), was a pilot project launched in 1995 in

13 LaRocque, Norman (2005) School Choice: Lessons from New Zealand, Briefing Paper No. 12, Education Forum, Wellington, www.educationforum.org.nz.

14 Patrinos, Harry Anthony (2005) Education Contracting: Scope of Future Research, PEPG Report 05–23, Program on Education Policy and Governance, Harvard University, Cambridge Massachusetts.

Participating private schools may be secular or religious and must be accredited.

‘‘ ‘‘

2www.cfbt.com 26


Quetta, the capital city of Balochistan. Its aim was to increase girls’ enrolments in school through the creation of private girls’ schools in poor urban neighbourhoods. Under the UGF, private schools were paid a three-year declining subsidy of Rs150 per month per girl in the first year, Rs135 per month in the second year, Rs100 per month in the third year and nothing thereafter. In addition, schools would receive an enrolment fee of 200 rupees per girl per year. The subsidy was limited to 100 girls, although additional girls could be enrolled, as could boys provided their numbers did not exceed the number of girls. The government subsidy was made directly to the school on behalf of families. Ten poor neighbourhoods were selected for the pilot.

3.5 Adopt-a-School Programmes

3.5.1 Sindh Education Foundation

70. The Sindh Education Foundation (SEF) operates an Adopt-a-School (AAS) programme in the Pakistan province of Sindh. Under the AAS programme, government schools are adopted by private sector individuals, companies or organisations. The AAS programme was launched in 1997. The programme aims to

improve government schools in the area of quality, access, infrastructure and community participation through:

• mobilisation of the private sector to assist public institutions;

• increased involvement of parents and communities through extensive community mobilisation;

• regular school monitoring and feedback;

• training headteachers and teachers; and

• increased co-curricular activities and improved school facilities.

71. The SEF plays the role of facilitator between the school and the adopting body and provides ongoing technical support and monitoring of processes and outcomes. To date, some 147 schools with 34,379 students have been adopted in 11 districts of Sindh. Seventy percent of AAS programme schools are in Karachi. Adopters include a range of individuals and organisations, including NGOs, Community Based Organisations, concerned citizens and the armed forces. Other AAS programmes exist in Pakistan, including one operated by the National Education Foundation and another operated by the PCP.

The subsidy was limited to 100 girls, although additional girls could be enrolled, as could boys provided their numbers did not exceed the number of girls.

‘‘ ‘‘ 140

120

100

80

60

40

20

0

16,000

14,000

12,000

10,000

8,000

6,000

4,000

2,000

–

Number of Schools Voucher Recipients

1990/

91

1991

/92

1992

/93

1993

/94

1994

/95

1995

/96

1996/

97

1997

/98

1998

/99

1999

/00

2000/

01

2001

/02

2002

/03

2003

/04

2004

/05

2005

/06

Num

ber

of S

cho

ols

Vo

uche

r R

ecip

ient

s

Source: http://dpi.wi.gov/sms/choice.html

FIGURE 7: Number of Participating Private Schools and Voucher Recipients, MCPC, 1990/91 to 2005/06

www.cfbt.com 27


3.5.2 Adopt-a-School Programme, The Philippines

72. The Adopt-a-School (AAS) programme was established in July 1997 through the enactment of Republic Act No 8525. However, the Implementing Rules and Regulations and the regulations authorising the tax incentives associated with the AAS programme were delayed for several years and were not approved until March 2003. The programme aims to:

• encourage the private sector, broadly defined, to partner with the Department of Education to address problems such as shortages of classrooms, desks and textbooks;

• provide mechanisms that allow the private sector to render assistance in upgrading and modernising the Philippine education system;

• provide an environment that is more conducive to learning, improve completion rates and lift achievement levels; and

• widen access to quality education and reduce student drop-out rates.

73. Under the programme, private entities are allowed to assist a public school, whether elementary, secondary or tertiary, preferably located in the twenty poorest provinces. This assistance can involve any number of activities, including staff and faculty development for training and further education, construction and upgrading of facilities, provision of books and other instructional materials and modernisation of instructional technologies. By 2006, some 22,000 schools had benefited from more than $50 million provided by 300 donors under the AAS programme.

3.6 Capacity Building Initiatives

74. Capacity building initiatives involve a range of activities:

• curriculum and pedagogical support;

• management and administrative training;

• textbook provision;

• teacher training; and

• development of support networks, professional partnerships and linkages.

3.6.1 Cluster Based Training of Teachers, Punjab (Pakistan)

75. In Punjab, the PEF operates a Cluster Based Training of Teachers through PPP (CBTT) programme. The CBTT programme provides professional development for private school teachers, with a focus on primary education. Training programmes focus on development of teachers’ knowledge of content, rather than on pedagogical approaches. The training is provided to clusters of schools. Generally speaking, each cluster is made up of 7–10 schools and 30–35 teachers. Teachers are paid an allowance to attend the training to cover transportation and other costs. Unit costs are around Rs1,250, although these can vary depending on the provider that is contracted to deliver the training or whether it is carried out by the PEF’s own staff.

3.6.2 Quality Advancement and Institutional Development in Private Sector Schools

76. Aga Khan Education Services Pakistan operates a programme entitled Quality Advancement and Institutional Development (QuAID) in Private Sector Schools. The overall objective of the project is to strengthen the capacity of low cost private schools so that the education they provide to poorer communities is of high quality and well managed. The focus is also on presenting sustainable models for delivering quality education. The QuAID programme works in a partnership with the Directorate of Private Education and private schools. It is currently operating in Karachi and Hyderabad, with 50 private schools in the former and 30 public schools/23 community schools in the latter.

3.6.3 Quality Assurance Resource Centre

77. The SEF operates a Quality Assurance Resource Centre (QARC), an educational development project aimed at developing a quality assurance certification programme as a means of influencing the quality of education at both public and private schools. The proposed certification programme would allow schools’ quality to be categorised as a means of informing parents’ schooling decisions. It would also provide for tailored quality enhancement support for public, private and community/

2www.cfbt.com 28


NGO schools, through for example, training and capacity building of teachers, school heads and managers. The programme has only just been developed. Its budget is nearly Rs40 million.

3.6.4 Teaching in Clusters by Subject Specialists

78. The PEF has recently begun piloting a programme called Teaching in Clusters by Subject Specialists. Under this programme, the PEF engages subject specialists to teach in a cluster of 3 schools (1 government/2 private). Subject specialists are usually in areas such as English or Science. Teachers are paid market rates – around Rs20,000 per month. This is a very small scale programme, with a small number of teachers being hired to date.

3.7 School Infrastructure Partnerships

79. PPPs are an increasingly common form of procurement for large infrastructure projects in the education sector. Infrastructure PPPs can be structured in a variety of ways. Under the most common type of PPP arrangement – Build-Operate-Transfer (BOT) – a private operator is granted a franchise (concession) to finance, build and operate an educational facility such as a public school, university building or hostel. The government, in effect, leases the facility from the private sector for a specified period, after which the facility is transferred to the government.

80. While arrangements can differ widely, infrastructural PPPs have a number of characteristics in common:

• private sector partners invest in school infrastructure and provide related non-core services (e.g. building maintenance);

• the government retains responsibility for the delivery of core services such as teaching;

• arrangements between the government and its private sector partner are governed by long-term contracts – usually 25–30 years. Contracts specify the services the private sector has to deliver and the standards that must be met;

• service contracts are often bundled, with the private sector taking on several functions such as design, building, maintenance and employment of non-core staff; and

• payments under the contract are contingent upon the private operator delivering services to an agreed performance standard.15

81. Infrastructure PPPs differ from traditional procurement methods in several ways. First, the private sector provides the capital required to finance the project. Second, the government specifies the contract in terms of ‘outputs’ or service level requirements, rather than in terms of ‘inputs’. Third, the newly constructed facility is not turned over to the government upon completion. As noted above, it is operated by the private sector until the end of the contract period. Several examples – drawn from the United Kingdom, Pakistan, Australia, Egypt, Germany, the Netherlands and Canada – are described briefly below.

3.7.1 Private Finance Initiative, United Kingdom

82. Private Finance Initiatives (PFIs) in the education sector have been used extensively in the United Kingdom. The PFI programme was introduced under the Conservative government in 1992 and has been strongly supported by the Labour Party since it took office in 1997. The government uses PFIs only where it is appropriate and where it expects them to deliver value for money.

83. Under a PFI programme, a capital project, such as a school, hospital or housing estate, is designed, built, financed and managed by a private sector consortium, under a contract that typically lasts for 30 years. Contracts can be structured differently. Under the Design-Build-Finance-Operate (DBFO) model, a private sector partner (usually a consortium of companies) takes on the provision and long-term operation of a facility in line with the local authority and school or schools’ specification. The private consortium is paid regularly from public money, based on its performance throughout the contract period. If the consortium misses performance targets, its payment is reduced.

15 Department of the Parliamentary Library (2002) Public Private Partnerships: An Introduction, Research Paper No. 1 2002–03, Commonwealth of Australia, Canberra, p. i.

PPPs are an increasingly common form of procurement for large infrastructure projects in the education sector.

‘‘ ‘‘

www.cfbt.com 29


At the end of the contract period, the school is returned to the government.

84. PFI uptake in the education sector was slow in the early years, but grew considerably following the introduction of a number of programme improvements since the late 1990s. As at October 2007, the Department for Children, Schools and Families (DCSF) had signed 115 PFI deals, with a value of approximately £4.8 billion. In addition, the Scottish and Welsh Governments had signed more than 20 education related PFI deals. The DCSF projects alone represented around 19 percent of signed PFI projects and 8.5 percent of the value of PFI projects. Among the largest education PFIs have been the South Lanarkshire and Glasgow Schools Projects in Scotland (£394 million and £206 million respectively) and the Northamptonshire Schools Project (£191 million) in England.16 Two UK studies found that PFI projects were more likely to be delivered on time and within budget than traditionally procured projects.17

3.7.2 School Private Finance Projects, Australia

85. The New Schools Project in the Australian state of New South Wales, consists of two main components. First, the private sector financed, designed and constructed nine new public schools in the state between 2002 and 2005. These new schools were built to standards that met or exceeded the Department of Education and Training (DET) school design standards. Second, the private sector will provide cleaning, maintenance, repair, security, safety, utility and related services for the buildings, furniture, fittings, equipment and grounds of these schools until 31 December 2032. In return, the private sector will receive performance-related monthly payments from the DET during the operational phase of the project. At the end of the contract period, the buildings will be transferred to the public sector.

86. The New Schools Project in New South Wales is part of a broader move toward PPPs

in Australia. PPPs have been used by various governments to procure infrastructure across a range of sectors, including transport, health and prisons. They have also been used in higher education, with the University of Southern Queensland and Swinburne University of Technology both using PFIs to construct educational infrastructure. The New South Wales government recently announced that the Axiom Education consortium will design, construct and maintain the state’s AUD$149 million New Schools Project 2 over a 30-year concession period. The project involves the construction of nine schools throughout regional and metropolitan New South Wales. A December 2005 New South Wales Treasury post-implementation review of the original New Schools Project found them to be positive and an improvement on traditional delivery.18

87. Three other Australian states are moving in the direction of school infrastructure PPPs. In South Australia, the New Schools Public Private Partnership Project is in its early stages. Under the programme, the private sector will finance, design and construct six public schools, accommodating some 4,000 students. The private sector will also provide cleaning, maintenance, repair, security and other services under long-term contracts with the State. The Queensland state government has announced plans to build seven schools under PPPs. Private operators would be responsible for the construction and maintenance of these schools over a 30-year contract, but education services would continue to be provided by the government. In Victoria, the government has announced plans to engage the private sector to build 11 schools and is looking at extending the PPP model to include fixing many existing schools.

3.7.3 PPP for New Schools, Egypt

88. The Government of Egypt has recently undertaken one of the largest PPPs in the education sector. Under the initiative, the government provides land, while the private sector partners design, construct, finance

16 http://www.hm-treasury.gov.uk/media/B/E/pfi_signeddeals_231007.xls17 HM Treasury (2003) PFI: Meeting the Investment Challenge, Her Majesty’s Stationery Office, London, p 43. 18 New South Wales Treasury (2005) New schools privately financed project: Post implementation review, Office of

Financial Management, Sydney, Australia, p. 2.

The Government of Egypt has recently undertaken one of the largest PPPs in the education sector.

‘‘ ‘‘

2www.cfbt.com 30


and furnish public schools and provide non-educational services under long term agreements. The initiative began in late 2006 and involved 300 schools in 23 Egyptian governorates. The private sector’s positive response to the initiative led to its expansion in 2007 so that it will now cover 2,210 primary and secondary schools, valued at some LE11 Billion. The first tranche of 345 schools in 18 governorates is currently being tendered.

3.7.4 Public-Private Partnerships for Educational Infrastructure, Nova Scotia (Canada)

89. The Province of Nova Scotia, Canada used a PPP (P3) model to build 39 schools in the late 1990s. The government pursued the P3 model because its financial situation was such that it could not afford to build the large number of schools required, especially given its desire to outfit them with state of the art technology. The first lease agreement between the government and private sector partner was signed in 1998.

90. Under the P3 model, schools were designed, built, financed and maintained by the private sector. Contracts were allocated on the basis of a competitive bidding process. The P3 schools were leased by the government for a period of 20 years. Most of these contracts expire between 2017 and 2020. Incentives were built into contracts to ensure quality construction and maintenance. Approximately 14 percent of the square footage in the province’s schools is found in P3 schools.

91. The government had planned to build 55 schools, but the number was cut back when the project was beset by a variety of political and other problems, including cost overruns driven by project ‘gold plating’ (that is, increasing school standards, expensive site selection) and weak bureaucratic management.19

3.7.5 School Infrastructure PPPs, Germany and The Netherlands

92. Although still in its infancy, there are some examples of infrastructure PPPs operating in the education sector in continental Europe. The Offenbach schools PPP project provides for the renovation, upkeep and facility management of over 90 schools within the County of Offenbach, which is located near Frankfurt, Germany. The project, which involves the government contracting for the financing, refurbishment and operation of government schools, is split into two parts, with a combined capital value of over EUR780 million. The first part of the PPP project involves 43 schools. The second part of the projects involves 49 schools. The private sector partners will operate schools for a period of 15 years.20 A schools PPP project in Cologne involves the refurbishment and operation of seven schools at five different locations in the City of Cologne, Germany. The value of the contract is EUR125 million. The company will operate the schools for a period of 25 years.21

93. The first Dutch education PPP project, which commenced in 2005, involves the construction and operation of a new secondary school in the Ypenburg suburb of The Hague. The secondary school is expected to grow from 150 students at the beginning of the contract to 1,200 by 2009. The Design-Build-Finance-Maintenance (DBFM) contract is for 30 years (1.5 years for construction and 28.5 years of maintenance). Maintenance will include cleaning, furniture, information and communication technology (ICT) infrastructure and possibly catering.

3.7.6 P3 New Schools Project, Alberta (Canada)

94. In June 2007, the Government of Alberta announced the P3 New Schools Project. Initially, the project involved 14 schools, but this was increased to 32 in January 2008.

19 Meek, Jim (2001) School’s Out, Summit Magazine, Volume 4, Issue 1, March, www.summitconnects.com.20 Freshfields Bruckhaus Deringer (2004) ‘Germany: Schools project sets benchmark’, Project Finance Monthly,

September, www.freshfields.com/practice/finance/publications/newsletters/pf-monthly/9366.pdf. Last accessed March 2006.

21 HOCHTIEF (2005) ‘HOCHTIEF celebrates building kick-off for Cologne schools PPP project’, Press Information, September 8.

Although still in its infancy, there are some examples of infrastructure PPPs operating in the education sector in continental Europe.

‘‘ ‘‘

www.cfbt.com 31


Under this project, a private-sector partner is responsible for the design, construction, finance and maintenance of schools for 30 years. Once the schools are open, government makes regular payments to the partner for 30 years. The government is guaranteed a fixed price and delivery date and receives a 30-year warranty on the schools. Risks such as construction cost inflation and weather-related delays are assumed by the private contractor. The contractor can be penalised for late delivery.

3.7.7 Leasing of Public School Buildings to Private operators, Pakistan

95. Another form of school contracting involves the leasing of under-used and dilapidated government school buildings to private schools. Under such arrangements,

the private sector is given the right to operate a school in the afternoon shift, when the school building is closed. In exchange, the private operator must upgrade the building, pay the utility costs of both schools, contribute to the operating costs of both schools and pay 10 percent of any profits to the public school council. The Government imposes a fee maximum of Rs200 per student per month at the primary level and Rs300 at the secondary level.

96. Such arrangements were initiated by the Punjab Government in February 2001 in four districts and the programme was extended to the entire province in May of the same year. More than 6,000 such schools are now operating in Punjab.

Under this project, a private-sector partner is responsible for the design, construction, finance and maintenance of schools for 30 years.

‘‘ ‘‘

2www.cfbt.com 32


97. As shown in Section 3, there is a wide range of PPPs being employed at the Basic Education level around the world. These PPPs come in many forms and cover a wide range of activities. They are found in both developed and developing countries. There are many reasons why governments are increasingly opting for PPPs to assist in meeting their policy objectives at the Basic Education level. In particular, well-designed PPPs can:

• increase the level of financial resources committed to Basic Education. Private sector philanthropic initiatives have the additional benefit that funding is generally more flexible than public sector funding, the bulk of which is committed to teacher salaries, particularly in developing countries;

• supplement government schools’ limited capacity to absorb growing numbers of children, thereby expanding access and helping to reduce class sizes in government schools;

• increase the level of private sector knowledge, skills and innovation – whether pedagogic, technical or management related – that may not be available in the education sector;

• allow government education authorities to focus on core functions such as policy and planning, curriculum development and quality assurance where they have a comparative advantage over the private sector, rather than devoting resources to areas where it may not have such an advantage;

• allow for much greater innovation in the delivery of education by focusing on the outputs and outcomes desired from an educational provider, rather than specifying how those outcomes should be achieved;

• allow governments to circumvent unnecessarily restrictive employment laws and outdated government pay scales that limit governments’ ability to hire appropriate staff and organise delivery in the manner required;

• introduce a longer time horizon into public-private relationships and better align the interests of the public and private sectors;

• reduce the politicisation of schooling and reduce the degree of corruption in the education sector;

• make the cost of services more transparent through the use of explicit contracts and improved costing mechanisms; and

• sharpen competitive pressures in the education sector, thus generating efficiency gains and spurring greater innovation in education delivery.

98. At the same time, the use of PPPs can have downsides. Opponents argue, for example, that they can represent a loss of control for education authorities and result in a loss of accountability to the public. PPPs in education have also been criticised on the grounds that they represent a first step toward full privatisation. Still others have argued that the benefits of choice and competition are not evenly dispersed and can lead to widening inequalities between rich and poor.

99. PPPs also involve more complex arrangements that require detailed policy design, as well as financial and contract management capability. Poorly designed contracts, including weak or inappropriate incentives, may expose the government to significant financial and performance risks. The development of policy, as well as the formulation and specification of provider contracts, can be complex and time-consuming – particularly for bureaucracies unfamiliar with an external, output-based contracting model. Contract monitoring and enforcement are not costless and pre-suppose a well-functioning legal framework for ensuring property rights and timely and fair dispute resolution. To be successful, they may require considerable capacity-building or contracting in of expertise to operate effectively.

4 PPPs in Education: Lessons Learned and Future Directions

PPPs also involve more complex arrangements that require detailed policy design, as well as financial and contract management capability.

‘‘ ‘‘

www.cfbt.com 33


100. Some forms of PPPs are well suited to the Basic Education sector because the sector is relatively stable and predictable, both in terms of demographics and technology, which provides for longer planning horizons. At the same time, however, PPPs face special challenges in the education sector, particularly in developing countries. These challenges include weak PPP implementation capacity in the education sector in many countries and the wide range of factors (particularly external ones) that affect school outcomes.

101. In addition, the fact that education is seen as ‘public’ or non-commercial in nature can limit governments’ interest in exploring PPPs even when these might lead to improvements in educational outcomes. It may also lead to reduced interest in the private sector since it means that policy reversals that may reduce the benefits arising from PPPs are more likely than in economic sectors.

4.1.1 Evidence on the Impact of PPPs in Basic Education

102. Although there is a wide range of PPPs in place around the world, rigorous evidence on the impact of these programmes is, for the most part, limited. According to Patrinos (2005) the best evaluations of programmes involve experiments that randomly assign benefits and include a true control group. In the absence of random design, or some form of natural experiment, rigorous techniques such as propensity score matching, local average treatment effects, regression discontinuities, and so on, are preferred. The nature of PPPs makes such projects highly amenable to proper impact evaluations. A recent World Bank (2006) review concluded that there is little rigorous research on the effectiveness of most types of education PPPs. As Patrinos (2005) notes, this is especially the true in the case of PPPs operating outside the United States and for non-voucher programmes. Selected research – at both a macro level and for individual programmes – on the impact of PPPs on student outcomes and other variables is

reviewed briefly below and summarised in Annex 2 below.

103. At a broad level, Woessmann (2005) examines the association between student achievement and education PPPs using student-level data for 35 countries drawn from the OECD’s Programme for International Student Assessment. He finds that, controlling for a range of student characteristics, public operation of schools is negatively associated with student performance in maths, reading and science, while public funding of schools is positively associated with student performance in these same subjects. This evidence suggests that school systems based on PPPs where the state finances schools but contracts their operation out to the private sector are more effective than either fully privately or publicly financed/delivered systems or systems involving significant private financing but public delivery. The advantage of private over public delivery is particularly strong in countries with large shares of public funding.22

Concession Schools104. There is some evidence that the Concession Schools programme has had positive impacts on schooling outcomes. Barrera (2005) examined the impact of Colombia’s Concession Schools on drop-out rates and test scores relative to those in public schools. He tested three hypotheses concerning the impact of concession schools on school outcomes and found strong evidence that Concession Schools have a direct effect on drop-out rates and some evidence that they have an impact on drop-out rates on nearby public schools. He also found weak evidence that Concession Schools have a positive impact on students’ test scores relative to those in public schools. His research showed that drop-out rates in Concession Schools were 1.7 points lower than in similar public schools, and that students in Concession Schools scored higher on mathematics (1 point) and language (2 points) compared to students in similar public schools.

22 Woessmann, Ludger (2005) Public-Private Partnerships in Schooling: Cross-Country Evidence on their Effectiveness in Providing Cognitive Skills, Research Report PEPG 05–09, Program on Education Policy and Governance, Harvard University.

…public operation of schools is negatively associated with student performance in maths, reading and science, while public funding of schools is positively associated with student performance in these same subjects.

‘‘

‘‘

2www.cfbt.com 34


PACES Voucher Scheme105. Several rigorous evaluations of the PACES voucher scheme in Colombia have shown the programme to be successful. According to this research, voucher students were more likely to attend a private school, more likely to complete 8th grade and their scores on standardised tests increased by two-tenths of a standard deviation (Angrist and others 2002). The unit cost of the programme was lower than that in the public sector. The programme also had longer-term positive effects, with lottery winners being more likely to take the college entrance exam (Angrist and others 2006). The empirical results point to an increase in (proxy) high school graduation rates of 5–7 percentage points, relative to a base rate of 25–30 percent. Tobit estimates place the treatment effects at around 2 points, or roughly .2 relative to the standard deviation of latent scores. The study also found that language scores increased by .33 standard deviations.

Fe y Alegria106. Swope and Latorre (2000) examined the impact of FyA schools in nine Latin American countries. They found that unit costs in these schools were higher than in public schools when the community contribution was factored in. However, they also found that:

• schools in the FyA network were successful in reducing repetition and drop-outs, with FyA schools in a majority of countries having lower repetition rates and definitive drop-out rates than other public schools (see Table 6); and

• progression rates and retention rates were 44 percent and 11 percent higher in FyA schools than in other public schools.

107. A more recent paper by Allcott and Ortega (2006) compares results on the Prueba de Aptitud Académica (PAA) for FyA graduates and control group of Venezuelan public school study calculates the Average Treatment Effect (ATEs) through Ordinary Least Squares (OLS) and propensity score matching. FyA students performed 0.05 and 0.06 standard deviations higher in verbal and math Scores, after correcting for observables (OLS results). The ATE for the verbal and math Scores were 0.09 and 0.14 standard deviations.

Urban Girls’ Fellowship Programme and Community Support Process Programme108. An impact evaluation of the Province of Balochistan’s Urban Girls’ Fellowship (UGF) Programme, which employed an experimental design, indicated that it increased girls’ enrolments by an average of 33 percentage

TABLE 6: Repetition Rates and Dropout Rates in FyA Schools and other Public Schools

Country Gross Repetition Rates(Percent)

Gross Dropout Rates(Percent)

Fe y Alegría Public Schools Fe y Alegría Public Schools

Peru 25.4 32.5 9.9 25.7

Bolivia 20.4 72.9 9.0 26.8

Venezuela 22.0 40.2 16.2 38.7

Nicaragua 24.1 39.1 13.8 10.0

Ecuador 7.2 12.8 29.2 38.6

Guatemala 20.5 18.0 22.3 38.2

Colombia 21.3 19.2 10.5 8.0

El Salvador 29.0 20.2 39.7 40.4

Paraguay 27.4 33.9 8.4 5.0

Source: Swope and Latorre (2000), pp. 104–105.

www.cfbt.com 35


points and boys’ enrolments by an average of 27.5 percentage points (Kim, Alderman and Orazem 1998a). The success of the programme varied across neighbourhoods, but was not linked to the relative wealth of a neighbourhood or to parents’ education levels. A similar programme in rural areas of Balochistan was less successful in that, while girls’ enrolments rose by 10.3 percent, boys’ enrolments fell by 6.8 percent (Alderman, Kim and Orazem 2003).

109. Another initiative by the Province of Baluchistan, the Community Support Process (CSP) programme, sought to encourage female enrolments by increasing the number of schools and female teachers, and encouraging parental involvement through the creation of community public girls’ schools in rural areas. The evaluation showed that the CSP programme increased girls’ enrolment by an average of 22 percentage points and boys’ enrolments by an average of 9 percentage points, even though boys’ schools had previously been available (Kim, Alderman and Orazem 1998b).

AVEC110. There is very little rigorous empirical evidence of the impacts of AVEC schools on education outcomes vis-à-vis regular public schools. A comparison of publicly subsidised AVEC schools and public schools shows that AVEC has a strong association with better schooling outcomes (Mora 2005). Repetition rates and drop-out rates are lower in AVEC schools than they are in public schools and AVEC schools have a higher percentage of teachers with higher education diplomas. In addition, AVEC schools have lower per student costs but a higher percentage of their budgets go to non-personnel costs.

Chile111. According to the Chilean national evaluation system (SIMCE), non-subsidised private schools perform best, followed by subsidised private schools, and then public schools (Bellei 2005, Mora 2005). However, among low income schools, public schools outperform private schools (Mora 2005). While a number of studies have attempted to measure the impacts of the voucher system on various outcome measures, the results are

contradictory and inconclusive (Patrinos 2005). Some found that the voucher system had positive impacts on test scores and pre-college examinations. Others found that there was no impact on test cores, repetition rates, or secondary school enrolment rates. Still others found that school choice had led to increased social and academic stratification. There are a number of reasons why it may be difficult to draw comparisons between public and private schools in Chile and how they explain the widely diverging results in individual analyses, all stemming from the lack of random assignment of students to schools (Bellei 2005).

Contract and Charter Schools112. There has been limited rigorous research undertaken into the effects of ‘contract’ schools in the United States and the United Kingdom. This research has been largely descriptive or qualitative in nature. A small number of qualitative studies have focused on the design and implementation of education contracting (see Hannaway (1999) and Bulkley et al (2004)). There is more evidence on the impact of Charter schools in the United States, with a number of studies using more sophisticated evaluation techniques. Even here, however, results have been inconclusive, with some studies indicating performance improvements, while others have shown either no performance improvement or deteriorating performance (Patrinos 2005).

Voucher Funding in Hungary and Czechoslovakia113. Filer and Münich (2000) examine the response of non-state and state schools to the introduction of a voucher funding system in Hungary and the Czech Republic in the early 1990s. The authors test three hypotheses relating to the voucher model: (i) whether non-state schools are established at locations where the supply of educational opportunities provided by state schools is low or of low quality, (ii) whether state and non-state schools respond to changes in the demand for education, and (iii) whether state schools respond to competition from non-state schools.

114. Their findings support the arguments advanced by voucher proponents, namely

There is more evidence on the impact of Charter schools in the United States, with a number of studies using more sophisticated evaluation techniques…

‘‘ ‘‘

2www.cfbt.com 36


that: (i) non-state schools emerge at locations with excess demand and lower quality state schools, and (ii) greater competition from non-state schools creates incentives for state schools to improve the quality of educational inputs used and significantly improve the quality of their graduates. They also find that non-state technical schools react to regional labour market conditions, but state schools do not.

School Infrastructure Partnerships115. Despite the growing prevalence of infrastructure PPPs, there remains little rigorous research on their impacts. Two studies by the Treasury and the National Audit Office (NAO) in the UK found that PFIs appear to have been successful in delivering public infrastructure in a timely and cost-effective manner. The Treasury study examined the UK’s experience in 61 PFI projects, while the NAO study examined 37 PFI projects. Both of these studies show that PFI projects were far more likely to be delivered on time and within budget than were traditionally procured (non-PFI) projects. For example:

• 88 percent of PFI projects examined in the NAO study and 76 percent of PFI projects examined in the Treasury study were delivered on time. This compares with just 27 percent of projects undertaken using traditional procurement methods. According to the Treasury study, PFI performance in the education sector was above average for PFIs, with 88 percent of PFI projects delivered on time; and

• 79 percent of PFI projects examined in the NAO study and 80 percent of PFI projects examined in the Treasury study were delivered within budget. This compares with just 27 percent of projects undertaken using traditional procurement methods. All changes to the contract price in PFI projects resulted from changes to user requirements (HM Treasury 2003: 46–48).

116. It is difficult to assess the operational performance of PFIs given that these contracts are up to 30 years in length and, as a result, contracts are still in the early stages. Nonetheless some initial evidence shows that PFI projects appear to be meeting expectations. For example:

• the 2003 Treasury study showed that 76 percent of public sector clients described private sector performance as ‘as expected’ or ‘better’. One-third of these described performance as ‘far surpassing’ initial expectations. One-quarter of those surveyed believed that private sector performance was either ‘less than expected’ or ‘much less than expected’;

• a 2001 NAO study found that 81 percent of public sector authorities who were interviewed were of the view that value for money from PFI projects was at least satisfactory. Only 19 percent felt it was either marginal or poor (HM Treasury 2003: 4).

117. The Treasury study also showed that there was considerable contestability in public procurement using the PFI process. Data on PFI competitions showed that PFI deals averaged four bidders each. The number of bidders was consistent across different sectors. Generally speaking, contract managers who were surveyed reported that price and design were the key factors affecting bidder selection. Procurement times averaged 23 months in the school sector, with a range of 15 to 25 months (HM Treasury 2003). (See Table 7.)

118. A value for money comparison prepared as part of the development of the ‘New Schools’ Private Finance Project in New South Wales, Australia also showed that PFIs can deliver more timely infrastructure at lower cost than traditional procurement methods. According to that analysis:

• schools built under the PFI model will be available an average of two years earlier than would have been possible using traditional public sector financing; and

• the net present cost to the Government of the schools over the next 30 years is expected to be reduced by approximately 7 percent compared to traditional procurement methods (Department of Education and Training 2003: 1–2).

119. A subsequent analysis came to a similarly positive conclusion (New South Wales Treasury 2005).

www.cfbt.com 37


4.1.2 The Role of Government in Education: Financing vs Provision

120. Several issues and lessons emerge from the review of PPPs in the Basic Education sector contained in Section 3 above. First, there is a wide range of initiatives that come under the PPP banner. These include PPPs defined in the traditional sense to include private philanthropic ventures such as school sponsorship, corporate and foundation donations and adopt-a-school programmes. In such cases, the private sector’s role is to provide finance to public schools, while the public sector retains responsibility for the delivery of education. A more recent trend has seen the emergence of a broader range of PPPs such as the private management of public schools, private school funding and voucher programmes and school infrastructure initiatives. These more recent PPPs represent a reversal of the traditional model of PPPs

described above in that they involve public sector financing and private sector delivery of Basic Education services. This involves a movement away from the top right-hand quadrant to the bottom left-hand quadrant in Table 8 below.

121. The distinction between government financing and government provision of education services is a key principle underlying this broader notion of PPPs and is based on an assessment of the proper role of government in education. It recognises that the government has a variety of policy instruments at its disposal to intervene in the education sector: funding, provision, regulation and the provision of information. Government intervention in education is typically justified on several grounds, including the fact that, left to itself, the private market would result in underinvestment in education

TABLE 8: Financing vs Provision in Government Services

FinancingProvision

Private Public

Private Traditional fee-paying private schools

Private philanthropic ventures

Tuition fees and other user fees in public schools

Adopt-a-school programmes

Public Contract schools

Charter schools

Voucher programmes

Private school subsidy programmes

Traditional public schools

TABLE 7: Price Certainty and Project Delivery, Traditional Procurement Method vs PFI, UK

Performance Indicator

PFI ExperiencePrevious (Non-PFI)

Experience(Percent)

NAO Study(Percent)

Treasury Study(Percent)

Construction projects delivered to the public sector early or on time

88 76 30

Construction projects where cost to the public sector was at or below the agreed contract price

79

80

27

Source: HM Treasury (2003), pp. 48–50.

2www.cfbt.com 38


because education may generate external benefits – i.e. benefits that accrue to the wider society not just to the individual undertaking the investment. However, as Poterba (1996) and Patrinos (2000) argue, the existence of externalities may justify a government role in financing, but not in delivering, education.23

122. Funding-based PPPs such as contract schools, Bogota’s Concession Schools, charter schools, voucher programmes and private school subsidy programmes are designed to bring together the strengths of public funding such as increased access to education with those of private delivery, including new skills, increased innovation and increased efficiency in delivery. Funding-based PPPs do not, as some critics argue, amount to the privatisation of education. Rather, they involve a redefinition of the role of government in education – from funder and provider to funder and regulator of education services.

123. Funding-based PPPs can also provide greater sustainability of education programmes and initiatives because the role of the private sector is mainstreamed into the policy itself. They can also provide for more rapid scaling up of successful initiatives in the Basic Education sector than would be possible under alternative models. A key constraint on growth in the private sector is families’ inability to pay fees. Subsidy-based PPPs recognise that there is, in many countries, a significant network of potential providers in the private sector and that this provides a ready means of expanding enrolments. Rapid increases in access to Basic Education are much more likely if through the traditional model of public financing and delivery – whether fully public, fully private or ‘traditional’ PPPs given that they can harness the full range of resources available, including infrastructure, a more flexible operating environment, management skills and teaching staff. Funding-based PPPs would also provide a catalyst for the expansion of the private school sector as new

schools spring up (or existing schools expand) in response to the increased demand for private education. Kim, Alderman and Orazem (1998b) find evidence that subsidies caused a significant increase in enrolments in poor urban areas (though not in poor rural areas).24 Similarly, Filer and Münich (2000) find that non-state schools established themselves at locations where there was excess demand and the quality of state schools was lower.

124. There are other examples of the ability of the private sector to scale up quickly. In the Punjab, the PEF’s FAS scheme has expanded from just 20,000 students in 54 schools in late 2005 to more than 270,000 students in over 670 schools today, with further expansion planned. Also in Pakistan, The Educators, a school franchise model operated by Beaconhouse Schools, has grown to 230 schools in 130 cities across Pakistan, with enrolments of 75,000. The World Bank’s Balochistan Education Project, a funding-based PPP, has seen 95 low-fee private schools established in its first year in response to a recently introduced subsidy programme. An additional 500 applications have already been received for the remaining 200 private school sites.

125. A re-orientation of government policy toward funding-based PPPs would provide much greater scope for scaling up programmes, thereby allowing for more rapid progress toward meeting EFA education goals. The PEF provides a good example of a government organisation that has built its strategy of improving education for the poor around funding-based PPPs (see Box 2).

4.1.3 Issues and Lessons for PPP Design

126. While PPPs can bring many benefits to the Basic Education sector, they must be done right if they are to succeed. As discussed above, poorly designed and implemented PPP programmes can expose governments to significant financial and policy risks. A number

23 Poterba, James (1996) Government Intervention in the Markets for Education and Health Care: How and Why?, in V. Fuchs (ed.) Individual and Social Responsibility: Child Care, Education, Medical Care and Long-term Care in America, National Bureau of Economic Research, University of Chicago Press and Patrinos, Harry (2000) ‘Market Forces in Education’ European Journal of Education, Vol 35(1).

24 Kim, Jooseop et al (1999) Can Private School Subsidies Increase Enrolment for the Poor? The Quetta Urban Fellowship Program, World Bank Economic Review, Volume 13, Number 3, pp. 443–465.

www.cfbt.com 39


of broad principles to guide the design and implementation of Basic Education PPPs are discussed below.

127. An important component of a PPP framework for Basic Education is an enabling policy and regulatory environment and a strong legal framework. The regulatory framework must create the conditions under which private firms can operate effectively and efficiently, while at the same time ensuring that the sector delivers high quality education and that the wider public interest is protected. Private education remains controversial. Its sustainability will depend crucially on its ability to deliver high quality teaching that produces the educational outcomes desired by parents and students. Market perceptions of the quality of private education are fundamental in this regard, and can be easily damaged. Bad publicity about some private providers who provide poor quality services can harm the reputation of the sector as a whole and affect its ability to attract students – a form of contagion. A strong regulatory and legal framework can help guard against such an outcome. Key features of such a regulatory and legal framework include ensuring that:

• entry requirements for new providers are: clear, objective and are not onerous (beyond obvious regulations aimed at assuring safety);

• there are no restrictions on providers’ organisational form (that is, allow both for-profit and not-for-profit providers to operate);

• education and other relevant legislation (e.g. employment laws) do not unduly restrict schools’ ability to operate effectively and efficiently;

• parents are provided with good information on private schools and their performance;

• there is a system of independent quality assurance that guards against low quality providers; and

• there is a range of interventions available to address situations where schools are not performing.

128. Active participation by the private sector in education is most likely to be encouraged if the government puts in place an appropriate legal framework to govern contract procurement and private sector investment more generally. This includes:

Box 2: The Punjab Education Foundation

The Punjab Education Foundation was established in 1991 and restructured in 2004 into an autonomous and independent institution with the overall objective of promoting quality education for the poor through partnerships with the private sector. The PEF is funded by the Government of the Punjab province of Pakistan. In 2005/06, the PEF budget stood at Rs73 million. The PEF is headed by a 15 member government appointed Board of Directors. A majority of these Directors, including the Chairman, are drawn from the private sector.

Since 2004, the PEF has introduced two funding-based PPP programmes aimed at increasing access to quality education for the poor. The PEF’s flagship programme is the Foundation Assisted Schools programme, which pays low-fee private schools a monthly subsidy of Rs300 per student enrolled. Participating private schools cannot charge tuition fees and must meet PEF quality standards in order to remain part of the FAS programme. The PEF has also introduced the Education Voucher Scheme (EVS), a small voucher programme that operates in a number of slum areas. There are currently 5,000 students involved in EVS.

The PEF has introduced two programmes aimed at improving teaching in low-fee private schools. The first of these, the Cluster Based Training (CBT) programme, provides cluster-based training for teachers in low-fee private schools. Training is aimed at building teachers’ content knowledge in mathematics, English and science. To date, 46,000 teachers have been trained under CBT. Under the second programme, Teaching in Clusters by Subject Specialists (TICSS), the PEF hires subject specialists to teach in a cluster of low-fee private schools and public schools. These teachers are paid well above prevailing salaries.

2www.cfbt.com 40


• introducing mechanisms to minimise the likelihood or appearance of corruption;

• reducing red tape and unnecessary regulation;

• assuring judicial independence, as well as timely and effective enforcement of contracts; and

• introducing policies and incentives that encourage private investment.

129. The government must also ensure that the grant of authority over education policy to municipalities and the division of responsibilities between the different levels of government is clear. This will provide greater certainty to both parties involved in the contracting relationship. The more enabling is the policy and regulatory environment and the stronger is the legal framework, the more likely it is that the government and potential private sector contractors will be able to arrive at terms and conditions that are mutually satisfactory and that make private investment in the education sector feasible and profitable.

130. The US experience with charter schools indicates that ‘strong’ charter school laws (that is, those that are favorable to the establishment and operation of charter schools) are correlated with both the number and viability of charter schools in each state. States with stronger charter school laws also have higher and more comprehensive student achievement than states with weaker laws. Based on several years’ analysis of

charter school laws in the USA, the Center for Education Reform has identified a number of desirable properties of charter school laws. These include not placing limits on the number of charter schools allowed in the State or district, allowing multiple avenues for authorising charter schools, providing charter schools with full funding and providing charter schools with considerable legal and operational flexibility (including exemptions from collective agreements and education laws).

Split the purchaser and provider roles within the education administrative body131. A better environment for PPPs can be created if the different functions within the education department are split so as to separate its purchaser and provider roles. Under a split purchaser/provider model, policy and regulatory functions would be separated from service delivery and compliance functions. Such a split would ensure that education purchase decisions at the local level are made in a more neutral manner vis-à-vis the public and private sectors. As long as the same ‘arms’ of government are responsible for both purchase and provision (and regulation), there is a risk that governments will favor delivery by the public sector over the private sector given that private sector competition can threaten the viability of public schools. As Snell (2002) argues ‘splitting policy functions from service delivery creates incentives for governments to become more discriminating consumers, looking beyond government monopoly providers to a wide range of public and private providers’ (Snell 2002).

Box 3: Principles to Guide the Development of a Private School Subsidy Scheme

Kardar (2001) notes four principles that should guide the development of such a private school subsidy scheme:

• the need to attract more schools into the formal system, by making the entry qualifications less stringent and the school registration procedures speedier and less cumbersome;

• ease pressure on the input costs of schools charging low fees, by focusing on inputs over which the federal and provincial governments have primary control;

• financial support should not be limited to one-time capital grants, but should be extended to cover recurrent spending over an extended period, subject to particular criteria being met; and

• implementation must take into account issues of public sector institutional capability and design must take into account the stock of available skills and the potential for abuse of discretionary powers.

Source: Kardar, Shahid (2001) Private Sector in Education, Systems (Private) Limited, Lahore

www.cfbt.com 41


132. In the USA, some states go further in their effort to split the purchaser and provider functions in education by allowing multiple authorisers for charter schools. For example, a group seeking to operate a charter school could seek approval to operate either from the local school district, a university or some other body.

Ensure the capacity of the contracting agency133. A key to successful design and implementation of PPPs is to ensure that the government agency that is responsible for PPPs has both the information and skills required to design, develop and manage the more complex contracting processes that underlie PPP programmes.

134. First, it is important that the contracting agency have good financial and administrative information systems. Good price, output and quality benchmarks are essential for the contracting agency to undertake an informed assessment of the bids submitted by organisations seeking to deliver education services. For example, any assessment of whether the bidding process is generating value for money requires that the contracting agency has good information on the unit cost of existing or alternative sources of provision – both in the public sector and in the private sector. It is also important that the contracting agency have good baseline information on education outcomes, both in general and in the schools to be contracted out, if it is to set appropriate performance benchmarks for private sector contractors.

135. Second, it is vital that the contracting agency employs people with the skills required to manage and oversee the complex task of contracting with private sector partners. While some people see the move away from public provision as government ‘withdrawal’ from education, it is not. Rather, it represents a shift in the role of government from provider of a service to facilitator and regulator. The implementation of contracting models and similar PPPs in education places new demands on the public sector and requires much different skill sets to implement than traditional methods of procurement.

136. In particular, the move from input controls to output-based contracting means that government agencies must develop their capacity to:

• assess the various services to determine when and under what circumstances contracting, rather than direct public provision is to be used;

• design, negotiate, implement and monitor education service contracts;

• develop enabling legislation that supports a competitive and transparent system of contracting; and

• develop appropriate quality assurance mechanisms.

137. Given the complex and multi-faceted nature of contracting, it is likely that a range of skill sets would be required in the contracting agency, including educational and pedagogical skills, contract management, economics and finance. A move to contracting for education services also requires that public officials adopt a different approach and a new administrative culture to what existed in the past. As Harding (2002: 22) has noted, in relation to health contracting (but which is equally applicable to education):

Contracting requires a drastic mind shift for public officials, from thinking of themselves as administrators and managers of public employees and other inputs, to thinking of themselves as contract managers with ultimate responsibility for delivering services.

138. The contracting authority must also ensure that it has the necessary payment and fraud monitoring systems in place to track payments and ensure that claims for payment from participating schools are legitimate and accurate. The payments system should also ensure that payments to schools are delivered in a timely fashion. The Philippines’ experience with ESC is instructive in this regard, as audits discovered some instances of fraud in the form of ‘ghost schools’ that were receiving funding under ESC, yet existed in name only. In addition, the payments system under ESC was not timely, so that payments to schools under the scheme were often delayed several months, a factor which discouraged many

…it is important that the contracting agency have good financial and administrative information systems.

‘‘ ‘‘

2www.cfbt.com 42


providers from participating in the ESC programme. An effective audit procedure is a necessary component of any payment/fraud monitoring system. NGOs can often be successfully employed in such roles.

139. A key element in ensuring capacity in the use of PPPs is the development of manuals, check lists, toolkits and standardised contracts to assist in the design and implementation of PPPs in the education sector. It should also involve the dissemination of case studies of PPPs, as well as the dissemination of best practice guidelines and lessons learned from the implementation of PPPs. The UK makes extensive use of such guidance materials (see http://www.teachernet.gov.uk/management/resourcesfinanceandbuilding/FSP/schoolsprivatefinanceinitiative/). Several other countries have developed infrastructure PPP guidance materials, including South Africa, India and Egypt. Centralised PPP units can help in this regard.

Employ a transparent, competitive and staged process for the selection of preferred providers140. A key element of effective contracting is that the bidding process should be transparent and competitive. Bidding for service delivery contracts should be open to all private organisations. This includes both for-profit and not-for-profit providers. Contracts should be open to local, national and international organisations who may wish to bid to operate a public school. The bidding process should be competitive whenever possible.

141. Schools whose management will be contracted out should be identified well in advance and the list should be made publicly available, perhaps through an easily accessible public register. The bidding process should also be set out clearly and in advance. A Request for Proposal (RFP) should be sent out to all potential bidders and publicised widely to ensure as broad a market as possible. The result of the bidding process should be advertised to ensure that market participants are made aware of the successful provider.

142. A transparent and competitive bidding process is likely to have positive effects in both the short and long term. In the short term, a

competitive bidding process is most likely to result in the bids that deliver value for money (that is, the lowest price for a given level of desired quality). They are also most likely to result in reduced corruption in contract awards. Over the longer-term, a competitive process is most likely to build market confidence in both the bidding process and the contracting agency, thereby helping to grow the private education services market over time.

143. It is important that the contracting agency implement a staged process for the selection of the preferred provider of education services. The process should include a number of steps:

• clarification of requirements, including development of contract objectives, as well as specification of desired services and expected outcomes;

• development of procurement strategy and identification of procurement team;

• development of the RFP associated with the contract;

• invitation of expressions of interest through the promulgation of the RFP;

• carry out contract pre-qualification process in which bids are assessed against requirements and select short list of bidders;

• interview short list of bidders, assess proposals in greater depth and negotiate contractual issues with short list of bidders;

• select preferred bidder and award contract;

• advertise result of selection process; and

• commencement of service (International Financial Services London 2001: 13)

144. Savas (2000) provides a comprehensive discussion of the steps involved in carrying out a competitive process for contracting for the delivery of public services (Savas 2000: 174–210).

Establish quality assurance/ monitoring processes145. The success of PPPs is likely to be enhanced if they are accompanied by independent quality assurance/monitoring mechanisms to evaluate the provider performance and programme outcomes.

A transparent and competitive bidding process is likely to have positive effects in both the short and long term.

‘‘ ‘‘

www.cfbt.com 43


This would ensure independent, unbiased assessments of PPP performance. Well-designed QA mechanisms can provide valuable information to consumers, providers and government officials on the progress and outcomes of PPP initiatives. There are many mechanisms used around the world to assure quality – both in the private and public sectors. These include testing services, accreditation and school review:

• CfBT Education Trust has carried out reviews of schools in Oman under contract to the Omani government – a similar role to that carried out by public sector organisations such as the Education Review Office in New Zealand and the Office of Standards in Education in the UK;

• Standard and Poor provides school evaluation services (SES) to school districts in the US. SES analyses academic, financial and demographic indicators and trends, provides benchmarks and presents its findings on the performance of schools;

• private sector organisations such as the Educational Testing Service, Pearson Educational and Kaplan in the USA and the Center for Educational Measurement in the Philippines provide testing and assessment services that help track school performance;

• the Philippine Accrediting Association of Schools, Colleges and Universities operates a private, voluntary system of accreditation for private schools and higher education institutions in the Philippines; and

• the SPEIP and QARC programmes in Pakistan’s Sindh province both focus on lifting the quality of education in low-fee private schools.

146. It is important that QA mechanisms are focused on improving the quality of education delivered and on lifting education outcomes. Too often, much of what passes as school ‘supervision’ involves compliance, red-tape and enforcement of rules that add little to a child’s educational experience. Unnecessary rules and regulations foster an environment in which corruption can be rife. Other mechanisms could be employed, including requiring private schools to display quality ratings.

147. In order to protect consumers, private schools that participate in the PEF’s FAS programme must display a sign stating that they are subsidised and do not charge tuition fees. The sign also includes PEF contact information, should parents have concerns they wish to bring to the attention of authorities.

Establish appropriate performance measures and include performance incentives and sanctions for non-performance in contracts148. The establishment of appropriate performance measures is a critical element in any contract design. Performance measures provide the basis for determining whether the service provider has met the agreed terms and conditions of the contract and may also play a role in determining the compensation to be paid to the contractor. The specification of performance measures becomes even more important in those cases where compensation is linked to the attainment of performance benchmarks.

149. The selected performance measures must be appropriate and must be in line with the desired outcomes being sought by the contracting authority. This is because the contractor’s behavior will largely be driven by what can measured and what is rewarded under the terms of the contract. In other words, the contracting authority will ‘get what it contracts for’. Performance indicators should be specified, to the extent possible, in terms of measurable outcomes (for example, learning gains, reading levels, test scores, reduced drop-out rates and reduced teacher/student absenteeism), rather than inputs (for example, hiring additional staff, increasing spending on particular activities).

150. The selection of performance measures and the standards to be attained must be approached carefully because it can introduce perverse incentives and lead to undesirable outcomes. For example:

• a strong focus on academic outcomes (for example, test scores) in contracts may ‘crowd out’ some of the focus on softer skills such as teamwork;

2www.cfbt.com 44


• an overly rigid focus on measurable outcomes may lead to too little attention being paid to outcomes that are desirable, but which cannot be measured and hence cannot be compensated; and

• a strong focus on external test scores may provide schools with an incentive to ‘cream skim’ by refusing entry to students who are not likely to be ‘strong performers’.

151. This is not to argue that performance measures should not be set or that they should not be backed up by financial incentives. Performance measures and financial incentives can help align the interests of the school with those of students and the government (that is, help overcome the ‘principal-agent’ problem). Appropriate incentives can also help to ensure that schools remain focused on the needs of students and keep abreast of changing demands in the marketplace. The contract specification phase can be crucial to the success of the exercise and so needs to be carried out carefully and by a multi-disciplinary team. Contract targets and expectations should be realistic and achievable.

152. The degree to which performance indicators can be specified will vary depending on the nature of the contract. Performance measures are far more likely to be specific in cases where the services being purchased are narrow in scope and easiest to measure (for example, remedial instruction, literacy programmes) than in situations where the services being purchased are broader in scope and harder to measure (for example, whole school management). In support of this, Hannaway (1999) points out contracts with Sylvan Learning, which provides narrowly focused remedial instruction in reading and mathematics, included much more specific performance indicators than did contracts signed with Edison Schools, which manages whole schools (Hannaway 1999: 6). Clearly the ability of the contract to specify detailed performance indicators depends to some degree on the complexity of the tasks to be contracted for.

153. Performance indicators can be specified in a variety of ways, be measured both qualitatively and quantitatively and can be reported at different intervals. For example,

quantitative indicators such as standardised test scores can be supplemented by more qualitative methods of assessing performance such as surveys with parents and teachers, site visits by third-party review and alternative assessment methods for determining progress in areas such as leadership development, the arts and character development. Many of the PPPs outlined in Section 3 above employed performance measures and tied compensation (or continued participation in programmes) to performance. For example:

• PFIs in the UK, Egypt, Europe and Australia include performance measures such as the proportion of time that the facility is available for use;

• continued participation in the PEF FAS private school subsidy programme is tied to the performance of students on a PEF administered standardised test; and

• indicators such as standardised test scores and drop-out rates are used to monitor provider performance in the Bogotá Concession Schools model.

154. In addition to establishing appropriate performance measures, well-designed education serviced contracts should include performance incentives and sanctions for non-performance (that is, link payment levels to that attainment of performance standards). There are many examples of PPPs, including PFIs, school contracting initiatives and private school subsidy programmes, where payments are tied to performance. Providers that deliver services on time, to the required quality or meet specified outcomes are rewarded by higher payments, while those who fail to do so are penalised – either through reduced payments or, in some cases, by having the agreement terminated.

155. The inclusion of performance incentives and sanctions as part of agreements underlying PPPs must be accompanied by an effective performance monitoring framework. Monitoring must focus both on fraud prevention and on ensuring that objectives and targets are being met (e.g. improved test scores, reduced drop-out rates, infrastructure availability, etc). This is especially true for complex PPPs such as PFIs and funding-based initiatives such as school management

Performance indicators can be specified in a variety of ways, be measured both qualitatively and quantitatively and can be reported at different intervals.

‘‘ ‘‘

www.cfbt.com 45


and school subsidy programmes. A particular risk for funding-based PPPs is the potential for inflation of enrolment figures or the creation of ‘ghost’ schools. Various PPP programmes have adopted strategies to address this risk. These include the introduction of school accreditation schemes, third-party validations of enrolments and requirements that schools allow government agencies free and open access to schools and school enrolment data. All can help to mitigate the risk of inflated enrolments in programmes where funding depends on enrolments. For example, more than 100 schools were dropped from the ESC programme in the Philippines over a three-year period as a result of a tightening of accreditation standards, a reform introduced, in part, to address concerns about ghost schools. In the case of PFIs, contracts generally include the creation of ‘help desks’ to record and address problems such as infrastructure unavailability.

Use ‘operational contracts’ to provide the most scope for successful contracting 156. A key component of successful contracting is that the government should employ operational contracts under which staff are selected, employed and paid by the private sector managers, rather than by the government. ‘Operational contracts’ are far superior to ‘management contracts’, because they provide the private sector with greater flexibility to redesign work processes, select appropriately skilled staff, pay the salaries required to attract good staff and dismiss non-performing staff. Use of management contracts or the imposition of restrictions on school operations (beyond minimal standards required to assure safety) can significantly hamper the private sector’s ability to determine appropriate resource allocations, to introduce management and pedagogical innovations and to improve the quality of education delivered at public schools.

157. Education service contracts that require private sector providers to hire existing staff, maintain existing staffing and pay levels, maintain union contracts for teachers all restrict providers’ ability to make productivity gains and introduce changes aimed at improving the quality of education at the school. Limiting private providers’ ability to

pay for non-performing staff, vary pay levels or provide performance-based pay would have a similar effect. Clearly some minimal standards may be required – for example safety checks on teaching staff, etc.

158. The ‘operational contract’ approach to contracting with private education providers amounts to a ‘fresh start’ for schools. Under this approach, the government would simply pay the private provider an amount per-student for school operation, plus a management fee and then allow the provider to make all operational decisions, including staffing decisions. The provider would employ all staff. This is particularly important where private providers are being contracted to turn around ‘failing’ schools, given that poor teaching is often a factor in poor school performance.

Allow maximum contract flexibility for providers159. The government’s role should be to spell out the desired outputs and performance standards, set penalties for failure to achieve and rewards for success and then leave providers to decide the best way of organising themselves to deliver the required outputs to the specified standard. Providers must be given as much management freedom as is feasible. The need for flexibility is especially true in the area of staffing and employment, but it is also relevant in other areas such as curriculum, budget allocation, etc.

160. Forcing providers to operate within the same restrictive regulatory framework that hobbles public schools would significantly reduce the potential gains from moving to a contracting model and limit the positive impact of competition in the sector. Indeed, one recent study found that more than two-thirds of US school district superintendents surveyed believed that reducing bureaucracy and increasing flexibility was very important as a way to improve public education (Belfield and Wooten 2003: 14).

161. Contracts with private providers must reflect the nature of the service being provided. They must also be designed so as to encourage private sector investment and interest, while at the same time ensuring that risks are covered and that contracts remain

2www.cfbt.com 46


contestable. Many PPPs have adopted relatively long-term contracts. For example, PFI contracts are generally 25–30 years, Bogota Concession School contracts are for 15 years and charter school contracts are for 3–5 years. Long-term contracts provide private partners with greater certainty and thus generate increased private sector interest. This is especially important given that private education remains controversial and there is potential for policy reversals. Longer-term contracts also provide contractors with more time to achieve contract objectives such as improving school performance – which can often take five or more years. Contracting agencies could opt for longer contracts with private firms managing public schools. Egypt’s infrastructure PPP contracts are for 15 years – up to two years for school construction and 13 years for school operation.

162. The risk with longer-term contracts is that they may blunt some of the impacts of competition and limit the gains from contracting. They may also lock in poorly designed features of contracts for extended periods. However, these costs need to be traded off against the benefits of increased interest and reduced uncertainty for providers. To offset some of these effects, provider contracts often include clauses that require

ongoing performance and allow recontracting at intermediate points.

Need for an effective communications strategy evaluation of programme outcomes163. A key requirement for the successful design and implementation of PPPs at any level of education is an effective communications and awareness strategy. Such a strategy should include increasing the profile for PPPs at the Basic Education level (objectives, international experience, etc) as a means of ensuring informed debate on their relevance to meeting education outcomes, educating stakeholders on the potential advantages and disadvantages of PPPs, managing public expectations, promoting best practice in the development and application of PPPs, providing an information resource on PPPs (e.g. through the internet and other means) for the government, potential providers, stakeholders and the wider public and highlighting lessons learned.

164. Implementation of Basic Education PPPs should be accompanied by a well-designed, rigorous evaluation. This would provide essential information on the success or otherwise of the programme and would expand the wider international knowledge

Source: Compiled from DFID/RSPN (2003) Multi-sector Dialogue on Private Public Partnerships for the Delivery of Basic Services in Pakistan

Box 4: Selected Characteristics of a Desirable PPP Framework

• Comprehensive in that it covers the range of functions, including financing, capacity building, human resources, accountability and transparency, equity and monitoring to ensure appropriate implementation

• Pro-poor in approach and be designed to recognise the varying levels of capability across districts

• Flexible enough to allow for different approaches across sectors and across districts

• Designed in consultation with partners and stakeholders, including community groups

• Interventions should be demand-driven, focused on beneficiaries’ needs and financially/politically sustainable

• Includes initiatives to build public sector capacity to implement PPPs

• Bureaucratic processes must be transparent

• Designed to ensure quality of delivery

• Framework should be institutionalised, with greater continuity of policy and programs and more transparent processes

• Include monitoring and evaluation framework.

www.cfbt.com 47


base on education PPPs. As Patrinos (2005) notes, too few rigorous evaluations of PPPs are available – especially beyond vouchers and outside the USA. Rigorous impact evaluations would increase the information base so that policymakers can make informed decisions.

4.1.4 Implications for Research and Parameters for Intervention Studies

165. One of the key messages from this review is that there is a wide range of emerging models through which the private sector can participate in the education sector. While the bulk of private involvement in education continues to be in the form of ‘traditional’ delivery of education in private schools – by NGOs, faith-based organisations, communities and commercially-oriented private entrepreneurs. The general expansion of the private Basic Education sector in many countries and the growth in PPPs open up new opportunities and pose new challenges for governments. In particular, private involvement in Basic Education and the development of PPPs may offer innovative ways of expanding access to education, especially for groups that have been poorly served by traditional delivery methods. At the same time, the trend toward private delivery poses particular challenges for governments – the need to design funding mechanisms that ensure equitable access to Basic Education and the need to develop systems to assure the quality of delivery in a decentralised education system.

166. The growth of the private sector and the emergence of a range of Basic Education PPPs both have implications for the role of international organisations and NGOs, as well as for private educational entrepreneurs. For the most part, education lending by international organisations such as the World Bank has mirrored the public sector dominance of Basic Education that exists in many countries. As Sosale (2000) has shown, World Bank lending for education amounted to $4.9 billion for the 1995–1997 period. Only about 11 out of 70 projects (around 15 percent) during that period included a

private sector component – and only about half of these projects were at the primary or secondary education level.25 It is likely that the expansion of PPPs in the education sector will lead to greater emphasis being placed on private sector components in project lending by international organisations such as the World Bank.

167. Increased emphasis on PPPs in international organisation projects could open up opportunities for international organisations, NGOs and the private sector to facilitate or undertake a much wider range of roles and functions in the Basic Education sector, in addition to ‘traditional’ roles such as operating schools and undertaking capacity building. These could include:

• managing public schools under contract to governments;

• offering school review services;

• establishing and operating school registration, accreditation, quality assurance and information systems;

• administering publicly financed voucher and scholarship programmes;

• offering policy advice to improve the regulatory framework for private Basic Education in client countries;

• facilitating the establishment of representative associations for private sector providers and mechanisms for coordinating public and private endeavours in the Basic Education sector;

• providing financing and business advisory facilities for private schools and;

• developing the market for infrastructure PPPs.

168. These activities would see private sector skills and resources brought to bear on a much wider range of activities – delivery of tuition and infrastructure services, business support and capacity-building, private sector financing, private sector regulation and information provision. Collectively, these activities would facilitate and support

25 Sosale (2000) Trends in Private Sector Development in World Bank Education Projects, The World Bank, Washington DC, p12.

2www.cfbt.com 48


the operation of a more decentralised and market-based education sector. They would also recognise that PPPs – in whatever form – generally place considerable demands on the capacity of both public sector organisations and private providers.

169. The expansion of PPPs has also highlighted a number of significant gaps in knowledge with respect to the private involvement in education. These gaps are evident at many levels and suggest several areas for future research. At a minimum, there is a need for much better data on the size and nature of the private Basic Education sector. Many developing countries lack even the most basic data on private Basic Education, including the number of schools, teachers and enrolments. In addition, not enough is known about the nature of private provision and the socio-economic status of the communities served by private schools. For example, despite recent evidence of the important role played by low-fee private schools in serving poor communities (Tooley 2005; Andrabi, Das and Khwaja 2006), a myth persists that private schooling serves primarily (or only) elites. In many countries and jurisdictions, information on private schooling is either non-existent or is collected only in a haphazard fashion. This provides little basis upon which to assess the potential effects of either funding and regulatory policy changes or assessing the potential role for PPPs in Basic Education. A key focus of future research efforts should be in documenting the scale and scope of the private school sector in developing countries. Such efforts should not be one off. Rather, governments should ensure that data collection on the private sector is built into the country or jurisdiction’s education management system and that such information is available in a format that is useful for policy development purposes.

170. The recent expansion of PPPs at the Basic Education level has also exposed a significant gap in our knowledge about the effects of different types of PPPs on education outcomes. As Patrinos (2005) and Barrera-Osorio and Patrinos (2009 forthcoming) have noted, there is an extensive literature on school choice initiatives in the United States, but a much less extensive

(though growing) international literature on school choice initiatives. There are few rigorous impact evaluations and even fewer random evaluations of voucher programmes – one of the few being the random evaluation carried out of the Colombia PACES programme. As Kremer (2003) shows, the scarcity of rigorous studies is especially true for developing countries.

171. A key focus of future research should be on financing and undertaking rigorous impact evaluations of existing and new PPPs at the Basic Education level. More and better programme evaluations would increase the information base so that policymakers can make informed decisions on the appropriateness of PPPs and their potential for improving education outcomes and attaining the MDGs. To address this gap, Patrinos (2005) has argued for a multi-country, multi-institutional, multi-year initiative involving a series of studies on choice initiatives. Use of state of the art programme evaluation techniques and a common methodological framework across a range of countries would ensure more consistent results and would provide information on the extent to which research findings can be generalised across countries.

172. Research should also go beyond simply quantifying the impact of Basic Education PPPs on education outcomes. It should also seek to determine why particular programmes are successful, including the conditions and circumstances under which particular types of PPPs work best. This is especially important given:

• countries’ vastly different cultural, economic and social circumstances;

• the significant variation in country capability; and

• the wide range of PPPs that are available to policymakers, including school contracting initiatives, voucher programmes, subsidy programmes, infrastructure PPPs and capacity building.

173. Research should focus on documenting regulatory frameworks for private Basic Education in different countries, as well as identifying and promulgating world best

The expansion of PPPs has also highlighted a number of significant gaps in knowledge with respect to the private involvement in education.

‘‘ ‘‘

www.cfbt.com 49


practice in regulatory and programme design. This could include developing mechanisms for information sharing (e.g. conferences, seminars, websites, etc), developing training modules and various other forms of capacity building, and preparation of toolkits for governments considering the introduction of PPPs. Key research areas include assessing existing regulatory frameworks and identifying legal and regulatory framework designs that promote the introduction and effective operation of PPPs, assessing government and private sector capability for introducing

PPPs, identifying training and capacity building needed to support the introduction of PPPs and designing information and quality assurance systems to complement the introduction of PPPs. Private involvement in Basic Education in general and PPPs in particular remain controversial. School choice and contracting programmes have faced considerable opposition in many countries. This argues for focusing research effort on both communications and implementation aspects of PPP policy reform to a much greater degree than has been done in the past.

2www.cfbt.com 50


174. This report has undertaken a review of the international experience with PPPs at the Basic Education level. These examples have been drawn from a wide range of both developed and developing countries. Several forms of PPP have been highlighted, including private philanthropic initiatives, private sector management initiatives, private school funding programmes (e.g. subsidies and vouchers), adopt-a-school programmes and school infrastructure partnerships. Private participation in education – and particularly Basic Education – remains controversial in many countries. Despite this, the number (and diversity) of PPPs at the Basic Education level is growing. ‘Traditional’ private philanthropic PPPs continue to play a significant role in financing education in many countries. More recently, many countries have introduced more sophisticated funding-based PPPs that involve government finance, but private sector delivery, of education services.

175. A clear message from this international review is that, contrary to a widely held view, Private participation in Basic Education does not need to favour those who are well off. Indeed, as the review shows, PPPs are often targeted on populations who are being poorly served by existing education delivery systems. This is the case with, for example, Concession Schools in Bogotá, vouchers programmes and contract schools in the USA and ESC in the Philippines.

176. Have PPPs been successful in the Basic Education sector? While they hold promise, it is too early to tell. To date, there has been limited rigorous research on the impact of various types of PPPs. While there is more – and more sophisticated – research on charter schools in the USA and vouchers, less is known about other examples around the world. Isolating the influence of PPPs on educational outcomes is not an easy task. The experience to date does suggest some tentative lessons for the design and implementation of PPPs at the Basic Education level and the conditions under which PPPs will work. A strong regulatory framework, flexibility in provision and good quality assurance are fundamental. More sophisticated PPPs such as school infrastructure initiatives and funding-based PPPs represent a significant design and implementation challenge for government departments. They require a redefinition of the role of public agencies and, often, a different set of skills on the part of the civil servants responsible.

177. PPPs in the Basic Education sector are clearly no panacea. Progress toward MDGs and improvements in education outcomes more generally will require much broader reform programmes. Nonetheless, PPPs – if done right – are a useful tool for governments to achieve their educational policy objectives. To be successful, they require good policy design, careful implementation and effective political management.

5 Conclusion

Have PPPs been successful in the Basic Education sector? While they hold promise, it is too early to tell.

‘‘ ‘‘

www.cfbt.com 51


Alderman, H., Kim, J., and Orazem, P. (2003) ‘Design, evaluation, and sustainability of private schools for the poor: the Pakistan urban and rural fellowship school experiments’, Economics of Education Review 22(3): 265–274.

Andrabi, T., Das, J. and Khwaja, A. I. (2006) A Dime a Day: The Possibilities and Limits of Private Schooling in Pakistan, Policy Research Working Paper 4066, World Bank, Washington DC.

Angrist, J., Bettinger, E., Bloom, E., King, E. and Kremer, M. (2002) ‘Vouchers for Private Schooling in Colombia: Evidence from a Randomized Natural Experiment’, American Economic Review 92(5): 1535–58.

Angrist, J., Bettinger, E. and Kremer, M. (2006) ‘Long-Term Educational Consequences of Secondary School Vouchers: Evidence from Administrative Records in Colombia’, American Economic Review 96(3): 847–862.

Barrera, F. (2005) ‘Impact of Private Provision of Public Education: Empirical Evidence from Bogotá’s Concessions Schools’, prepared for the conference ‘Mobilizing the Private Sector for Public Education’, Harvard University, 5–6 October.

Barrera-Osorio, F. and Patrinos, H. A. (2009 forthcoming) ‘An International Perspective on School Vouchers’, Handbook of Research on School Choice, Routledge.

Belfield, C. and Wooten, A. (2003) Education Privatization: The Attitudes and Experiences of Superintendents, Occasional Paper No. 70, National Center for the Study of Privatization in Education, Columbia University, New York.

Bellei, C. (2005) ‘The Private-Public School Controversy: The Case of Chile’, presented at the Conference on ‘Mobilizing the Private Sector for Public Education’, Harvard University, 5–6 October.

Bettinger, E. (2005) Lessons from Private-School Vouchers in Colombia, Federal Reserve Bank of Cleveland, Cleveland OH, www.clevelandfed.org/research/EdConf2005/Papers/BettingerPaperFINALweb.pdf

Bulkley, K. et al (2004) Contracting Out Schools: The First Year of the Philadelphia Diverse Provider Model, Research Brief, Research for Action, Philadelphia.

Center for Education Reform (2004a) Strong Laws Produce Better Results, CER Special Report, Washington DC.

Chakrabarti, R. and Peterson, P.E. (forthcoming) School Choice International. MIT Press: Cambridge MA.

Department of Education and Training (2003) New Schools Privately Financed Project: Summary of Contracts, Government of New South Wales, Sydney.

Department of the Parliamentary Library (2002) Public Private Partnerships: An Introduction, Research Paper No. 1 2002–03, Commonwealth of Australia, Canberra.

Department for International Development/Rural Support Programme Network (2003) Multi-sector Dialogue on Private Public Partnerships for the Delivery of Basic Services in Pakistan.

Education Review Office (2004) Alternative Education Report, Wellington

Filer, R. and Münich, D. (2000) ‘Responses of Private and Public Schools to Voucher Funding: The Czech and Hungarian Experience’, William Davidson Institute Working Paper No. 360, Ann Arbor.

Foundation Center (2007) Foundation Giving Trends: Preview. New York.

Foundation Center (2007) Highlights of Foundation Yearbook: 2007 Edition. New York.

Gaffney, M. and Smith, A. B. (2001) ‘New Zealand’s Targeted Individual Entitlement Scheme’, in Can the Market Save Our Schools? Fraser Institute, Vancouver: 151–166.

Hannaway, J. (1999) Contracting as a Mechanism for Managing Education Services, Consortium for Policy Research in Education, Policy Brief RB–28.

Harding, A. L. (2002) ‘Introduction to the Private Participation in Health Services Handbook’, in April L. Harding and Alexander

6 Bibliography

2www.cfbt.com 52


S. Preker, Private Participation in Health Services Handbook, The World Bank, Washington DC: 7–74.

HM Treasury (2003) PFI: Meeting the Investment Challenge, Her Majesty’s Stationery Office, London.

International Financial Services London (2001) Public Private Partnerships: UK Expertise for International Markets, London.

Kardar, S. (2001) Private Sector in Education, Systems (Private) Limited, Lahore.

Kim, J. et al (1999) ‘Can Private School Subsidies Increase Enrollment for the Poor? The Quetta Urban Fellowship Program’, World Bank Economic Review, Vol. 13, No. 3: 443–465.

Kim, J., Alderman, H. and Orazem, P. (1998a) Can Cultural Barriers Be Overcome in Girls’ Schooling? The Community Support Program in Rural Balochistan, Working Paper Series on Impact Evaluation of Education Reforms Paper No. 10, World Bank, Washington, DC.

Kim, J., Alderman, H., and Orazem, P. (1998b) Can Private Schools Subsidies Increase Schooling for the Poor?: The Quetta Urban Fellowship Program, Working Paper Series on Impact Evaluation of Education Reforms Paper No. 11, World Bank, Washington, DC.

Kremer, M. (2003) ‘Randomized Evaluations of Educational Programs in Developing Countries: Some Lessons’, American Economic Review 93(2): 102–106.

LaRocque, N. (2006) Contracting for the Delivery of Education Services: A Typology and International Examples, Education Forum, Wellington, www.educationforum.org.nz.

LaRocque, N. (2005) School Choice: Lessons from New Zealand, Briefing Paper No. 12, Education Forum, Wellington, www.educationforum.org.nz.

Lozada-Crystal, D. (2001) RVR-AIM Report on Corporate Giving Philippines 2001, Manila.

McMeekin, Robert W (2003) Networks of Schools, Education Policy Analysis Archives, Arizona State University, Tempe, http://epaa.asu.edu/epaa/v11n16.

Meek, Jim (2001) ‘School’s Out’, Summit Magazine, Vol. 4, Issue 1, www.summitconnects.com, March.

Molnar, A. et al (2006) Profiles of For-Profit Education Management Organizations: 2005–2006, Eighth Annual Report, Education Policy Studies Laboratory, Arizona State University, Tempe, AZ.

Mora, J. (2005) ‘Public-Private Partnerships in Latin America: A Review Based on Four Case Studies’, prepared for the conference ‘Mobilizing the Private Sector for Public Education’, Harvard University, 5–6 October.

New South Wales Treasury (2005) New schools privately financed project: Post implementation review, Office of Financial Management, Sydney, Australia.

Swope, J. and Latorre, M. (2000) Fe y Alegría schools in Latin America: Educational communities where the pavement ends, Centro de Investigación y Desarrollo de la Educación, Santiago: 104–105.

Patrinos, H. A. (2005) Education Contracting: Scope of Future Research, Program on Education Policy and Governance Report 05–23, Harvard University, Cambridge MA.

Patrinos, H. A. (2000) ‘Market Forces in Education’, European Journal of Education, 35(1):61–80.

Patrinos, H. A. and LaRocque, N. (2007) Enhancing Accountability in Schools: What Can Choice and Contracting Contribute, World Bank, Washington, DC.

Patrinos, H.A. and Sosale, S. (2007) Mobilizing the Private Sector for Public Education: A View from the Trenches, World Bank, Washington, DC.

Poterba, J. (1996) ‘Government Intervention in the Markets for Education and Health Care: How and Why?’, in Fuchs, V. (ed.) Individual and Social Responsibility: Child Care, Education, Medical Care and Long-term Care in America, National Bureau of Economic Research, University of Chicago Press.

Rodriguez, A. and Hovde, K. (2002) The Challenge of School Autonomy: Supporting Principals, LCSHD Paper Series No. 77, World Bank, Washington DC.

Savas, E. S. (2000) Privatization and Public-Private Partnerships, Chatham House Publishers, N.Y.

www.cfbt.com 53


Snell, L. (2002) Charter Schools as School Privatization, Reason Public Policy Institute, 4 September, www.rppi.org.

Sosale, S. (2000) Trends in Private Sector Development in World Bank Education Projects, Policy Research Working Paper 2452, World Bank, Washington DC.

Srivastava, P. (2007) Neither Voice nor Loyalty: School Choice and the Low-Fee Private Sector in India, Research Publications Series, Occasional Paper No. 134, National Center for the Study of Privatisation in Education, New York.

Tooley, J. (2005) ‘Private Schools for the Poor’, Education Next, Winter: 22–32.

United Nations (2005) Progress Towards the Millennium Development Goals, 1990–2005, http://mdgs.un.org/unsd/mdg/Host.aspx?Content=Products/Progress2005.htm.

Vanourek, G. (2005) State of the Charter School Movement 2005: Trends, Issues, and Indicators, Charter School Leadership Council, Washington DC.

Wang, Y. (1999) ‘Public-Private Partnerships in Health and Education: Conceptual Issues and Options’, prepared for ‘Manila Social Policy Forum: The New Social Agenda for East, Southeast and Central Asia’, Joint ADB-World Bank Conference, 9–12 November.

Woessmann, L. (2005) Public-Private Partnerships in Schooling: Cross-Country Evidence on their Effectiveness in Providing Cognitive Skills, Research Report PEPG 05–09, Program on Education Policy and Governance, Harvard University.

World Bank (2005) Colombia: Contracting Education Services, Report No. 31841-CO, Washington DC.

2www.cfbt.com 54


Annex 1: Summary of Selected International Examples of PPPs at the Basic Education Level

Programme Jurisdiction Programme Size Key Elements

Private Management of Public Schools

Contract schools United States • 521 EMO managed schools with 237,000 students in 2005/06

• School districts or charter boards contract with private providers to manage public schools

• Providers are paid a fee to operate the schools and schools remain free to students

• Focus on low-performing schools and school districts

Charter schools United States • More than 4,000 schools with over 1 million students in 40 states and DC in 2007/08

• Charter schools operate with fewer regulations than standard state schools, but must meet increased accountability requirements

• Schools remain free to students• Schools may be community managed or management may be

contracted out to for-profit or not-for-profit school managers

Bogotá Concession Schools

Colombia • 25 schools with 26,000 students

• 10 organisations managing schools

• Private schools and/or education organisations bid in competitive process for management contracts of newly built schools in poor neighbourhoods

• Contractors may manage a single school or a group of schools• Management contracts are for 15 years and are subject to

satisfactory performance• Schools receive Col$1,114,500 per full-time student per year• Schools operate with the flexibility and autonomy of private

providers

Independent Schools Qatar • 64 schools in 2007/08 • Government-funded schools with more operational autonomy/flexibility than public schools

• Run by government-selected not-for-profit operators – three-year contracts

• Enrolment-based funding

Management of Government Schools

City District Government of Lahore, Pakistan

• 97,000 students in 172 schools

• Management of government schools by Cooperation for Advancement Rehabilitation and Education (CARE), a local NGO

• Government pays salaries of its own teachers, while CARE employs its own teachers

• As part of the school management contract, CARE has invested around Rs600,000 in infrastructure in each school

• CARE has no administrative control over government staff – they remain government employees

Purchase of Educational Services from Private Schools

Government sponsorship of students in private schools

Côte d’Ivoire • 162,000 students in mid 1990s

• In 1995/96, 40 percent of private school students were state sponsored

• Budget of US$10.3 million in 1997–98

• Government contracts for the purchase of places in private secondary schools

• Payment varies with student’s educational level: US$200/year for lower secondary and US$233/year for upper secondary

• Schools must be ‘chartered’ and must maintain academic standards to be eligible to enrol sponsored students

Page 1 of 4

www.cfbt.com 55



Educational Service Contracting

Philippines • 383,000 students in 1,800 schools in ESC in 2005/06

• Budget in 2004/05 was $US26.6 million

• Under ECS, the government purchases places for students in private schools

• Programme operated by Fund for Assistance to Private Education (FAPE)

• Schools must be certified by FAPE to participate• Schools are paid up to PhP4,000 per student• Programme is targeted at low-income families

Financial Assistance Per Child Enrolled Basis (FAS)

Pakistan • Punjab Education Foundation initiative

• 1,157 schools and more than 500,000 students

• Subsidy paid to private school on a per-child enrolled basis • Schools’ students must meet minimum performance

benchmark for the school to remain eligible for funding• Professional development of partner FAS schools

Universal Post Primary Education and Training

Uganda • 430 private schools participating with enrolments of 56,000 students

• Only private secondary schools in sub-counties that do not have a government-aided school are eligible to participate

• Policy is targeted at low-fee schools – i.e. those which were charging no more than Ush75,000 per student per term

• Participating private secondary schools are paid a subsidy of Ush47,000 per student per term under the policy

• Participating schools chosen by Ministry of Education and Sports• Memorandum of Understanding signed with individual private

schools

Voucher Programmes

PACES Colombia • 125,000 vouchers over 1992–1997 period

• Operated from 1992–1997• Voucher open to secondary students from poor families

wishing to attend private schools• Voucher value = $US190 (just over half the cost of a private school)

• Voucher deposited by student; school receives funds from bank• Top-up fees allowed• Vouchers renewable subject to satisfactory academic

performance• Minimal regulation of private schools

School Funding System

Netherlands • 67 percent of total enrolments are in private schools

• All primary and secondary-school students are entitled to voucher

• Voucher can be used at all schools – public, private, religious, secular

• Weighted formula to benefit low-income groups and minorities• Schools reimbursed on the basis of attendance• Top-up fees not allowed• Extensive regulation of private schools (e.g. not-for-profit only)

Targeted Individual Entitlement

New Zealand • 160 vouchers per year • Established 1996 and new intakes ceased in 2000• Voucher scheme available to children from low-income families

to allow them to attend • Voucher value set at 110 percent of average cost of state

education • Vouchers could only be used at registered independent schools • Voucher programme was administered by organisation

representing independent schools

Page 2 of 4

2www.cfbt.com 56



Milwaukee Parental Choice Program

Milwaukee, USA

• 22,500 vouchers per year from 2006/07

• 121 participating schools

• Established 1990• Allows poor families to send their children to private or

parochial schools at state expense• Voucher value up to $US6,351/student• Participating private schools must be accredited • Number of vouchers equal to over one-quarter of all students

in Milwaukee school district

Urban Girls’ Fellowship Program

Quetta, Pakistan

• Implemented in 10 poor neighbourhoods of Quetta

• Limited to 100 subsidised girls per school

• Subsidies paid to on behalf of families in order to attract a private girls’ school

• Subsidy of Rs150/girl/month in 1st year up to max. of Rs10,000/school; lower subsidies in later years; schools received Rs200/girl for start up

Adopt-a-School Programmes

Adopt-a-School Programme

Sindh, Pakistan • 147 schools• 34,379 students

• Established 1997• Operated by the Sindh Education Foundation• 11 districts of Sindh• 70 percent of schools are in Karachi• Private individuals and organisations adopt government

schools

Adopt-a-School Programme

The Philippines • $4.9 million raised between 2002 and 2004

• Gives private institutions and individuals the opportunity to become partners in education by providing assistance in the upgrading and modernisation of public schools

• Established 1997

• Includes tax incentives for adopters

Capacity Building Initiatives

Cluster Based Training of Teachers through PPP

Punjab, Pakistan

• Operated by the Punjab Education Foundation

• 36 training programmes and 1,800 teachers trained

• Professional development of private school teachers – focus on primary education

• Schools in cluster cannot charge more than Rs400/month• Cluster = 7–10 schools and 30–35 teachers

Quality Assurance Resource Center (QARC)

Sindh, Pakistan • Rs39.98 million• Established by Sindh

Education Foundation

• Operated by Aga Khan Foundation

• Educational development project aimed at categorising and providing quality enhancement support for public, private and community/NGO schools

• Training and capacity building of teachers, school heads and managers

Quality Advancement and Institutional Development (QuAID)

Sindh, Pakistan • 103 schools in Karachi and Hyderabad

• 50 private schools in Karachi and 30 public schools/23 community schools in Hyderabad

• Focus is on building capacity of partner schools and presenting sustainable models for delivering quality education

Teaching in Clusters by Subject Specialists

Pakistan • Small number of teachers hired so far

• Operated by Punjab Education Foundation• Pilot project• PEF engages subject specialists to teach in a cluster of 3 schools (1 government/2 private)

• Teachers are paid market rates

Page 3 of 4

www.cfbt.com 57


Page 4 of 4


School Infrastructure Partnerships

Private Finance Initiative

United Kingdom • By the end of 2003, 102 education PFI deals had been signed, with a value of £2.028 billion.

• Introduced in 1992, but grew significantly post-1997• Educational infrastructure designed, built, financed and

managed by a private sector consortium, under a contract that typically lasts for 30 years. Design, build, finance and operate is most common structure

• Payments to private sector are performance related• Most new educational facilities are now built using PFIs

‘New Schools’ Private Finance Project

Australia • Nine new public schools in New South Wales built from 2002–2005. Nine more built from 2006

• Six new public schools in South Australia with 4,000 students

• Eleven schools planned in Victoria

• Seven schools planned in Queensland

• Private sector finance, design and construction of public schools

• Private sector cleaning, maintenance, repair, security, safety, utility and related services for school buildings, furniture, etc

• Long-term contracts

PPP for New Schools Egypt • 2,210 new primary and secondary schools in Egypt

• The first tranche of 345 schools in 18 governorates is currently being tendered

• Initial project started in late 2006 – 300 schools in 23 governorates

• Response by private sector led to expansion in early 2007 to include a further 1,910 schools around the country

• Government provides land, while private sector designs, constructs, finances and furnishes schools and provides non-educational services under 15 year agreements

Public-Private Partnerships for Educational Infrastructure

Nova Scotia, Canada

• 39 schools built under P3 programme in late 1990s.

• Schools are financed, built and operated by the private sector• Government leases schools for 20 years• Incentives built in to contract to ensure quality construction

and maintenance

Offenbach Schools Project and Cologne Schools Project

Germany • 92 schools in Offenbach County with capital value of over EUR780 million

• 7 schools in Cologne with value of EUR125 million

• Government contracting for the finance, renovation and operation of public schools in Offenbach County. Private sector partners will operate schools for 15 years

• Refurbishment and operation of schools in Cologne. Private sector will operate the schools for 25 years

Leasing of Public School Buildings to Private Operators

Punjab, Pakistan

• 3,000 schools • Began as pilot in 4 schools in February 2001• Private sector can operate a school for the afternoon shift,

when public school is closed• Private operator must upgrade building, pay utility costs and

contribute to the operating costs of both schools and pay 10 percent of any profits to the public school council

• Private school fee limited to Rs200/student per month at the primary level and Rs300 at the secondary level

Source: LaRocque, Norman (2006) Contracting for the Delivery of Education Services: A Typology and International Examples, Education Forum, Wellington, Patrinos (2006) and www.publicservice.co.uk/pdf/pfi/issue50/PJ50%20Kogan%20Pillay%20Atl.pdf

2www.cfbt.com 58


Page 1 of 1

Annex 2: Summary of Selected Evidence on Education PPPs

Programme/Location

Programme Details Study Evaluation Methodology

Key Findings

Plan de Ampliación de Cobertura de la Educación Secundaria (PACES), Colombia

Private school voucher

125,000 students in programme in 1997

Per-student cost = $US190, versus $350 per-pupil spending on government schools

Angrist and others (2002)

Angrist and others (2005)

Randomised evaluations with vouchers allocated by lottery

Lottery winners were 15–20 percent more likely to attend private school, 10 percent more likely to complete 8th grade, scored 0.2 standard deviations higher on standardised tests (equivalent to a full year); programme effects are larger for girls

Lottery winners were about 5–7 percentage points more likely than lottery losers to take the college entrance exam (20 percent increase); also had higher language scores than lottery losers, had improved test scores at mean and for the highest achievers

Urban Girls’ Fellowship Program, Quetta

Private school subsidy

Subsidy = Rs100/girl/month in 1st year up to max. of Rs10,000/school; lower subsidies in later years; schools received Rs200/girl for start up

Kim and others (1998a)

Randomised assignment using treatment and control groups

Fellowship programme positively affected enrolments for girls and boys.

Girls’ enrolments grew by an average of 33 points, while that of boys grew by an average of 27.5 points

Bogotá Concession Schools

Contract schools

25,000 students; per-student cost = Col$1.2 million (<average cost for public school students)

Barrera (2005) Propensity and matching estimation

Concession schools had lower drop-out rates (1.7 points) and higher mathematics (1 point) and language (2 points) scores than similar public schools

Fe y Alegría Contract schools

Unit costs in FyA schools exceed the cost of public schools

Allcott and Ortega (2006)

Ordinary Least Squares (OLS) and propensity score matching

FyA students performed 0.05 and 0.06 standard deviations higher in verbal & math, correcting for observables; ATE for verbal and math 0.09 & 0.14 standard deviations

Rural Girls’ Fellowship Program, Balochistan

Rural community schools

Implemented in 27 rural villages

Subsidy of Rs100/girl/month up to Rs2,400/school/month

Schools were offered start-up packages of textbooks and supplies

Kim and others (2003)

Randomised assignment using treatment and control groups

Girls’ enrolments rose by 10.3 percent, and boys’ enrolments fell by 6.8 percent

Czech Republic and Hungary

Voucher programme

Non-state secondary education grew from zero to 25 percent of schools (13 percent of enrolments) in the Czech Republic and 15 percent of schools (8 percent of enrolments) in Hungary, by the mid-1990s

Filer and Münich (2000)

Standard probit model, TOBIT and OLS

Non-state schools emerge where there is excess demand and lower quality state schools

Greater competition from non-state schools creates incentives for state schools to improve the quality of educational inputs and significantly improve the quality of their graduates

Source: Patrinos, Harry Anthony and Norman LaRocque (2007) Contracting for the Delivery of Education Services: A Typology and International Examples, Education Forum, Wellington, Patrinos (2006) and www.publicservice.co.uk/pdf/pfi/issue50/PJ50%20Kogan%20Pillay%20Atl.pdf

www.cfbt.com 59


Further research publications

Through the Evidence for Education programme, CfBT Education Trust is proud to reinvest its surpluses in research and development both in the UK and overseas.

Our aim is to provide direct impact on beneficiaries, via educational practitioners and policy makers. We provide a range of publications from practice-based intervention studies to policy-forming perspective papers, literature reviews and guidance materials.

In addition to this publication the following research may also be of interest:

• Supporting national educational reform in Sub-Saharan Africa, Michael Latham, Susy Ndaruhutse, Dr Harvey Smith, CfBT Education Trust

• Grade repetition in primary schools in Sub-Saharan Africa, Susy Ndaruhutse, CfBT Education Trust

• Private schools for the poor: a case study from India, James Tooley and Pauline Dixon

Forthcoming publications

Spring 2008

• Base-line primary education research in Angola, Professor Lynn Davies Education Action International

• New directions in schools provision in Africa, Dr Cilla Ross, Co-operative College

• Advocacy-driven educational programming in emergencies and early reconstruction a case study: making safe spaces work for children, Jonathan Penson and Kathryn Tomlinson

Summer 2008

• Alternative education programmes for refugee, internally displaced and returnee children and youth, Pamela Baxter and Lynne Bethke

• Protecting positive community participation in education during emergencies and reconstruction, Joan Sullivan-Owomoyela and Laura Brannelly

• Donor engagement in education during and after conflicts, Susy Ndaruhutse, Carole Rigaud and Laura Brannelly

For further information or for copies of the above research please visit our website at www.cfbt.com/evidenceforeducation or contact our Research and Development team at [email protected].

CfBT Education Trust

CfBT Education Trust60 Queens RoadReadingBerkshireRG1 4BS

0118 902 1000www.cfbt.com P

MS

402

0 05

/08

I

SB

N 9

78-0

-861

60-0

55-7

Date post:	29-May-2018
Category:	Documents
Upload:	hangoc
View:	212 times
Download:	0 times

LITERATURE REVIEW Public-Private Partnerships in … in... · 5 Public-Private Partnerships in...

Documents